Bulletin of the
British Museum (Natural History)

Spectacles and Silver Ys: a synthesis of the
systematics, cladistics and biology of the
Plusiinae (Lepidoptera: Noctuidae)

Ian J. Kitching

Entomology series
Vol54 No2 28 May 1987

The Bulletin of the British Museum (Natural History), instituted in 1949, is issued in four
scientific series, Botany, Entomology, Geology (incorporating Mineralogy) and Zoology,
and an Historical series.

Papers in the Bulletin are primarily the results of research carried out on the unique and
ever-growing collections of the Museum, both by the scientific staff of the Museum and by
specialists from elsewhere who make use of the Museum’s resources. Many of the papers are
works of reference that will remain indispensable for years to come.

Parts are published at irregular intervals as they become ready, each is complete in itself,
available separately, and individually priced. Volumes contain about 300 pages and several
volumes may appear within a calendar year. Subscriptions may be placed for one or more of
the series on either an Annual or Per Volume basis. Prices vary according to the contents of
the individual parts. Orders and enquiries should be sent to:

Publications Sales,
British Museum (Natural History),
Cromwell Road,
London SW7 5BD,
England.

World List abbreviation: Bull. Br. Mus. nat. Hist. (Ent.)

© Trustees of the British Museum (Natural History), 1987

The Entomology series is produced under the general editorship of the
Keeper of Entomology: Laurence A. Mound
Assistant Editor: W. Gerald Tremewan

ISBN 0 565 06025 2

ISSN 0524-6431 Entomology series
Vol 54 No 2 pp 75-261

British Museum (Natural History)

Cromwell Road

London SW7 5BD Issued 28 May 1987

Spectacles and Silver Ys: a synthesis of the
systematics, cladistics and biology of the Plusiinae
(Lepidoptera: Noctuidae)

Ian J. Kitching

Department of Entomology, British Museum (Natural History), Cromwell Road, London
SW7 5BD

Contents
RMR Reece te tft 3 IES es das aR gill a RE ATG es da bi ith nl cept AA EES 9 hase Shan q5
PEC MUON LECOCIICIUSH §.t25 Aras astctondyd apechere eet ubrdaed «hina Sesdacesedd Cacao chided deceit Peebles oe 76
TIRE CCM CLL ORIYA Teeth e Pe ease cee s tiene ekp fac kemd ate k « sbateu es va. faBids sdb sna Seemae rama see ks.» sie 76
Piahistory of the subfanuly Plusiaiae |i... 2fc020 60 cated. pot ecs vnc sencabedeeesdeslewaveccnenues 77
ie iuchenclassification oftheplustimesiy. i605. «0: csc: tacuies +t newhiaeaaneeas eleouls daie'dan Saclane 77
‘he position ofthe Plustinae within the Noctuidae ..5 5 icshiws.s. wcesaecwn assets se -aaes 92
WANS X ALDI Chait. ma taslkeuyieinacth. couch? bol dadsk tach agcechra ees artemis ascent chiaer ene 93
TET ino (0 CAV ener ase ae isc nadine poBE BE Meiricige ie TASC ASC OLY or Oore Cree Cepner EERE eer ne ae 94
ICHEGNAS Aah Ste tebe. Neri foams dip: SM Latah POE ds oe Ieee « caches gaaaeet 97
PeLEALACLGES fet neeeritferacsanay Baa tase Jaen t est... ones. egatebioda th ast 6b Likewise es 97
EMA NCA LOGE UMER es Hstet eth ake ne tens. Nac talederaiesce as, SERED. Wo soamacaels AE neath aac be 112
irsiG45-stenicladooiamismmeees A nt pencpewats: thc. od meee eben setae Bites Mere cee 6 113
ieeezAh een cna Tice mh P Lngai apnea pets ta seta sax Gein aaa 3 ded Ra Tp dw a Wad Re wees Ae vine 113
SE EZ UPLSTG TG mirc iat era 3. tet 0 toed adie Pete cela faci tad> sae gest hs poh stnagh Stands bacon 4ay~ Rhos +4 115
Se AUtognapia., C OnnutipluSia ang Uelisieyss tk eee em Pon scchana ocho iene odoin « sons ener 115
OL ALTOYOVAY RUE UOT sg ARs aOC en Oe Re ey CCR COREE SRR SCE. OPE ICRED NE ne Semana a DE 117
5. Rachiplusia and the branching sequence of the tribes and subtribes .................. 117
Si. UUIGSWC Uo cade i Bhat arte tetrad a tacit ema: Seater AAR OAD Se lr Er Aa aaa etna 119
CSTE LAC OULAMISt ern areeet etter e ca tancan cease ros crea enecensianhmoee erie ert ete encndanes 121
RAWAL OM Of CE PLELEMMEG CLAGOPTAM oe i sceccyccucavesccnescsedsesccebertcastarcecdseetencareces 121
(CIMEIAGIGT GIES Se sarge Une gee ceches eanbaser Scan dHae Sep cn eadeT ee iasn rap eee ey mamerr hn AROAae 124
CIRCA ITER ITS loneeliesetuadret dene bcorceinaGhh tied eben: ie Sel Goes eT eaRe Bee CARER Sanetiee 138
Mirek CUELLO CUT ALWSE Sy eer eee or Ge era Seine ots ada ean NAS acai anaee sn ctosin oe 167
WBOMISETISUIS!CLASSILICATION) “ccs wats sn acs Ma ORAM REE TNE wack on oc gcersienciencioe sinnicaieer.cisis oar 170
AAD SEI OVO LEE HUIS p ee tan com aeeRne ARBRE Hep odegee nae ces oae Ghieip Sr aaen her dodo auedeepnner caeene 170
PRCSULUG Eten curt eee tee scn terete as erste terckeater scene «ante eetnateme re randecsinederanctee 170
The positions of Shensiplusia, Desertoplusia and Pseudochalcia ..............6.0.0000004+ 171
ine higher classificationrof the Plustinae ii..)) 000). 1.05.. A Rn ere die cece 174
Comparison with previous classifications and phylogenies ..................0ccceeeee ee eens 175
POS CUS SLO MY BE BIEI Na ese Te ah ASE aI BR ethics CE A ARs Ras ARES S wateBAt WIL
Magvalitomdplantiass © ClathOmSifacti.rjv cite uit st wataadh Mitodge Ake Sea. eee eballas eee cece 177
2 CREO RE alah MR ee ae ee Eon treat treet res crcee cn nice oer. 178
ROUT NESTON docile EE eae ete ahi and iad ca De Set acepe pel Duron endo hares faa iat asictse fiche 179
PMLLETNG CAAT © CES ACA LGC LIYE OOO ete ete ceca a a Ne aes sssaaiin ase acetate piensa eeesaintie/ao Baecwssatse 180
[RGU STRELA SS ye eteetaeienad aaoe acre erie cc HGR “SRI TOI SOE OE EE a em 180
Appendix 1. Genera and type species of Plusiinae and outgroups examined ............. 187
Appendix 2. Plusiinae generic-level taxa not employed in this study ....................... 190
ANS OSE NUON Dera jae | DEEN 00a ooo. GuemoroenchoncringMBeRUcCpaOnEY poder eager ps (ance cr ae eenenne IIL
Appendix 4. New taxa and synonymy, status changes and revivals .....................005- 195
AppendixS}) Geneneclaxsification of the Plusiinae ..1....i2i0...00s ceiceedede eee ekcenetece en 195
[Noles caer casdeoamarcucccodei: conor occrcadmenen veneer OnGne HaCERESEE Cina. Gate cnet ME ECE: MEAT cee eenerecee 259
Synopsis

A comprehensive historical review of the higher classification of the Plusiinae is presented, with particular
reference to the different generic concepts and interrelationships that have been proposed. The position of

Bull. Br. Mus. nat. Hist. (Ent.) 54 (2): 75-261 Issued 28 May 1987

76 : IAN J. KITCHING

the subfamily within the Noctuidae is considered separately. A cladistic analysis of 47 generic-level taxa of
plusiines and 11 outgroups was performed, using the type species of each as an exemplar. A total of 216
characters is described, drawn from the morphology of the head, prothorax, vestiture, wings, legs,
abdomen and male and female genitalia. These data were analysed using the numerical cladistic program
PAUP and at least 50 equally-parsimonious cladograms of 648 steps were generated. The character state
changes supporting the various alternative subtopologies are discussed and a preferred 648-step cladogram
constructed. A further analysis revealed the existence of 16 647-step cladograms but, after character
analysis, these were all rejected. The preferred 648-step cladogram is evaluated, first in terms of
characters, and then in terms of its constituent clades. Under each of the genera in the latter section, a brief
review of biology, distribution and other points of interest is also included. This assessment allowed
characters of low reliability to be identified and subsequently eliminated from the data. The reduced data
set, less the outgroups, was analysed once more using PAUP. At least 50 equally-parsimonious
cladograms were again found. A manual strict consensus cladogram analysis was then performed using
these cladograms and, from the results, a consensus classification of the Plusiinae constructed, using the
conventions for annotated Linnaean hierarchies. Four tribes are recognized in the Plusiinae: the Omorphi-
nini, Abrostolini, Argyrogrammatini and Plusiini. The last is divided into three subtribes: the Euchalciina,
Autoplusiina and Plusiina. Several genera (Shensiplusia, Pseudochalcia, Diachrysia, Anagrapha, Rachi-
plusia, Loboplusia and Erythroplusia) cannot be placed with confidence and are included incertae sedis in
the classification. Cornutiplusia, Lophoplusia, Ctenoplusia and Thysanoplusia are recognized as good
genera, although with reservations regarding the last of these. Pseudoplusia, Eutheiaplusia, Acanthoplu-
sia, Adeva and Caloplusia are interpreted as subgenera of Chrysodeixis, Plusiotricha, Ctenoplusia,
Euchalcia and Syngrapha respectively. This classification is compared with the five recent higher
classifications proposed for the Plusiinae. Finally, the larval foodplant associations and biogeography of
the subfamily are briefly discussed. One new subtribe is erected and five new combinations made. One
species and one generic synonymy are proposed. Four previously synonymized genera are recognized as
valid while a further five genera are reduced to subgeneric status.

Acknowledgements

This study was carried out during the tenure of a Junior Research Fellowship awarded by the Trustees of
the British Museum (Natural History), which is gratefully acknowledged. I thank my colleagues at the
British Museum (Natural History) for their support and comments, the staff of the Electron Microscope
Unit, BMNH, for their assistance, and the staff of the Photographic Unit, BMNH, for the photographs
reproduced in this paper. I also thank Dr R. de Jong, Rijksmuseum van Natuurlijke Historie, Leiden, and
Dr W. Dierl, Zoologische Staatssammlung, Munich, for the loan of type specimens.

Introduction

For over 80 years, the higher classification of the family Noctuidae has consisted of a
non-hierarchical series of largely artificial subfamilies, based upon a small number of simple and
readily observable characters. There have been frequent small adjustments to the positions of
individual genera; more rarely a new tribe has been erected (although usually defined only for a
regional fauna), and never has a whole subfamily been revised. However, if a useful and
predictive higher classification of the noctuids is to be constructed, then problems have to be
addressed at the tribal level and above.

Subsequent to my recent review of noctuid higher classification (Kitching, 1984), I decided to
analyse the interrelationships of the genera of the subfamily Plusiinae, by the critical application
of cladistic methodology. The plusiines were chosen for a number of reasons. First, and perhaps
foremost, there were reasonable grounds for supposing that, even as currently constituted, the
Plusiinae are a monophyletic group. Thus, it would be unlikely that any of the constituent genera
would be overlooked by their being placed in other subfamilies. This supposition is strongly
supported by the observation that, since Hampson’s (1913) Catalogue, no pre-existing genera
have been transferred into the Plusiinae from elsewhere in the Noctuidae. Additional genera
have arisen either by the discovery of new species (e.g. Loboplusia, Eutheiaplusia) or by the
subdivision of pre-existing plusiine genera. A few genera (e.g. Phyprosopus Grote, Diloba)
have been transferred out of the Plusiinae. Second, the plusiines are a relatively small subfamily
of around 400 species and 45 genera. They were thus more amenable to analysis within my terms
of reference than, say, the more economically important but much larger Noctuinae (with

SYSTEMATICS OF THE PLUSIINAE Ti

approximately 2000 species in 150 genera). This is not to say that the plusiines are merely of
academic interest. The subfamily includes a number of major agricultural pests, notably
Trichoplusia ni, Pseudoplusia includens, Chrysodeixis spp. and Thysanoplusia orichalcea
(Fabricius). Thus, a predictive higher classification of the Plusiinae is vital for the effective
applied entomology of the group. Finally, the collections of the British Museum (Natural
History) (BMNH) contain specimens of about 80% of the world species and 96% of the genera
of plusiines: relatively complete coverage by any standards.

It was impossible to conduct a species-level analysis of the subfamily and so some selection of
taxa to be included had to be performed initially. I decided to concentrate on one species from
each genus or genus-group taxon, choosing the type species for convenience. Characters were
drawn primarily from the external morphology of the adult moths. Little consideration was
taken of the internal morphology, and none of the immature stages or biology. A cladistic
method was chosen primarily because of the explicit manner in which it handles characters and
their state transformations. There is no excuse for hiding behind vague or authoritarian
statements of relationships. I hope my reasoning on such issues as character coding, polarity and
interpretation are clear: that of many previous authors has not been so.

The history of the subfamily Plusiinae
The higher classification of the plusiines

In the following historical review of the Plusiinae, several terminological conventions have been
employed. Members of the subfamily in general are referred to as ‘plusiines’, while ‘plusias’
refers collectively to the tribes Argyrogrammatini and Plusiini (i.e. the Plusiinae less the
Abrostolini and Omorphinini). The taxon names used are those employed in the publication
being discussed at that point. Current nomenclature, where this differs, is given in square
brackets. This approach was adopted to facilitate back-referencing by the reader and to clarify
some of the earlier nomenclatural anarchy.

The emergence of the Plusiinae

Linnaeus (1758) described the first plusiine noctuids within the subsection of Phalaena Noctua
characterized as ‘spirilingues dorso cristato’. Of these, triplasia was well separated from the
remaining five species (chrysitis, gamma, interrogationis, jota Linnaeus and festucae). This
arrangement was followed by Denis & Schiffermiiller (1775), who included triplasia (together
with asclepiadis [Denis & Schiffermiller] and two other noctuids) in the informal group Y
(Larvae Pseudogeometrae, Ph[alaenae] Noctuae Submetallicae), and the other plusiines in
group Z (Larvae Semigeometrae, Ph[alaenae] Noctuae Metallicae). These groups were
accepted by Borkhausen (1792) and formally named by Ochsenheimer (1816) as the genera
Abrostola and Plusia. Haworth (1809), however, chose to treat triplasia and asclepiadis as a
section of Noctua (19 Arctuo-strigatae) and included the other plusiines in the genus Phytometra
(section Metallicae), a decision that was to have important repercussions subsequently.

Hiibner [1821] increased the number of plusiine genera to ten. Abrostola was again placed in a
separate ‘family’, while the remaining plusiine species were allocated to Euchalcia, Agrapha,
Syngrapha, Autographa, Polychrysia, Panchrysia, Diachrysia, Chrysaspidia and Chrysodeixis.
However, these genera were, by and large, ignored for the remainder of the nineteenth century.

The Plusiinae as a family-group entity originated as the Tribus Plusidi, including three genera:
Abrostola, Chrysoptera |Lamprotes| and Plusia (Boisduval, 1828). Duponchel (1844) modified
this arrangement by treating the latter two groups as ‘subgenera’ of Plusia. Plusia (s. str.) was
further subdivided into three unnamed sections based on wing pattern. Guenée (1841; 1852-4)
and Herrich-Schaffer (1845) proceeded along similar lines.

This bigeneric system persisted for many years (e.g. Grote, 1874; 1875; Hampson, 1893-5),
although Smith (1882-3) united the two into a single genus Plusia. Gradually, however, opinion
swung towards the view that the omnibus genus Plusia ought to be subdivided, given the
structural diversity displayed by its included species. Both Grote (1882) and Tutt (1891-2)

78 IAN J. KITCHING

considered Deva [Pseudeva] to be a good genus, and by 1875, Grote had accepted Hiibner’s
[1821] generic groupings, to which he added Chrysanympha. Butler (1892) also argued for
separate generic status for Polychrysia and considered Deva (sensu Grote and others) to be
synonymous. Polychrysia thus included additional species currently referable to Lamprotes and
Pseudeva.

On the basis of characters drawn from the wing-shape, labial palps and eyes, Dyar (1902)
recognized seven genera of plusias: Polychrysia, Panchrysia, Plusia, Autographa, Euchalcia,
Syngrapha and a new genus, Eosphoropteryx. This arrangement was endorsed, for the most
part, by Ottolengui (1902) in a paper published in the same journal only a few pages before that
of Dyar. Forbes (1914) reverted to the omnibus Plusia but included Dyar’s subdivisions in a
note.

Meanwhile, Hampson (1913) was subdividing the plusiines along completely different lines.
The subfamily was renamed the Phytometrinae, with the type genus and species, Phytometra
festucae. This change resulted from Hampson’s criterion for type species designation. Irrespec-
tive of previous designations, Hampson chose as the type species of a genus, the first species
included when that genus was described, unless otherwise explicitly specified by the author at
the time (McDunnough, 1916). Thus, the type species of Phytometra Haworth was considered to
be festucae, despite Westwood’s (1840) designation of aenea ([Denis & Schiffermiller]) [=
viridaria (Clerck)]. Thus Plusia, the type species of which was taken by Hampson to be deaurata,
became a subjective synonym of Phytometra.

In addition to ‘Phytometra’, fourteen other genera were placed in the ‘Phytometrinae’, on the
basis that they were quadrifine noctuids with lashed eyes. In his ‘phylogeny’ of the subfamily,
Hampson considered that the unknown ancestor had given rise to Omorphina (transferred to
the Erastriinae by Kostrowicki, 1961; but see below), Diloba (of current uncertain position, but
probably a noctuid), Abrostola and Palaeoplusia. Mouralia was derived from Abrostola, while
Palaeoplusia was ancestral to ‘Phytometra’. This last genus contained by far the majority of the
plusiines but from it arose five further genera: Plusidia, Pseudeva, Eosphoropteryx, Chrysoptera
[Lamprotes, Polychrysia] and Syngrapha. Plusiopalpa was considered to be a derivative form of
Chrysoptera, while, by additional tibial spination on the fore- and mid-legs and a reduction in the
size of the eyes, Autographa (comprising only the present Syngrapha parilis (Hubner), another
result of Hampson’s type species designations), Rachiplusia and Caloplusia were derived from
Syngrapha. Syngrapha itself was divided into two parts, corresponding approximately to the
current genera Syngrapha and Autoplusia. In Chrysoptera, sica (Graeser) was distinguished
from the remaining species by an excised forewing termen, while ‘Phytometra’ was split into four
groups, depending on whether the male antennae were bipectinate and apically ciliate (‘P.’
pterylota), bipectinate and apically serrate (‘P.’ psectrocera Hampson), serrate (‘P.’ lavendula
(Hampson) and ‘P.’ gerda (Piingerler)) or ciliate (the remaining 151 species). Chrysanympha
formosa was placed in Abrostola.

Hampson’s arrangement was followed in its entirety by the authors of ‘Seitz’: Warren (1913),
Draudt (1940) and Gaede (1937-9). However, other workers were far from satisfied. McDun-
nough (1916) corrected many of Hampson’s erroneous type species designations, including that
of Phytometra. This allowed him to replace ‘Phytometra’ with Autographa, return parilis to
Syngrapha and to reinstate Plusiinae as the subfamily name. At this time, it should be noted, the
type species of Plusia was considered to be chrysitis (Linnaeus), a point to which I shall return
below.

McDunnough’s revision of the Nearctic plusias

Barnes & McDunnough (1917) advocated a system that was something of a hybrid between
those of Dyar (1902) and Hampson (1913), and which was largely accepted by McDunnough
(1938) in his later checklist. However, shortly afterwards, McDunnough (1944) published a
revision of the Nearctic plusias, which represented a significant departure from all previous
classifications, and for the first time in plusiine systematics, genitalic characters were employed.
The subfamily (less Abrostola and Mouralia) was split into three sections based on the form of

SYSTEMATICS OF THE PLUSIINAE 79

the clavus. Section A, in which the clavus was a short, slightly setose knob arising from a broad
base, included Syngrapha, Anagrapha and Caloplusia, and was additionally characterized by
‘readily recognizable’ tibial spines. Section C comprised Pseudeva and Adeva, genera in which
the clavus was not clearly defined, while section B, in which the clavus was a long, thin, variably
setose rod, contained the balance of the species. This latter section comprised twelve genera:
Autoplusia, Trichoplusia, Argyrogramma, Pseudoplusia, Rachiplusia, Chrysaspidia [Plusia],
Autographa, Chrysanympha, Eosphoropteryx, Polychrysia, Plusia [Diachrysia] and Agrapha
[Allagrapha]. Palaeoplusia was synonymized with Chrysaspidia [Plusia].

By way of an aside, it was subsequently noted by Franclemont (1964) that Grote (1896) had
designated Phalfaena] Noct[{ua] glauca (Cramer) [= ahenea Hibner] as the type species of
Agrapha. Agrapha originally contained two species, aerea and glauca, which Franclemont
considered were not congeneric. Because he agreed with McDunnough that aerea represented a
genus distinct from Plusia [Diachrysia] but that Agrapha was inappropriate, Franclemont
proposed the genus Allagrapha, noting similarities in the male and female genitalia with
Macdunnoughia that were not shared with Plusia [Diachrysia].

Forbes (1954) followed McDunnough’s (1944) subdivision of the plusiines but treated the
latter’s genera as subgenera of Plusia, as did Zimmerman (1958). Inoue & Sugi (1958) attempted
to apply the new system to the Japanese fauna but the result was not very satisfactory;
Autographa, in particular, became highly heterogeneous.

Meanwhile, the studies of Forbes (1910), Fracker (1915), Gerasimov (1935) and Hinton
(1946), amongst others, were stimulating interest in the use of the detailed larval morphology in
lepidopteran classification. Initially, largely due to lack of material, the results consisted mainly
of descriptions (e.g. Gardner, 1941; 1947; 1948a) but soon sufficient taxa had been examined
and described to allow some comparative work to be undertaken. Crumb (1934) split the
plusiine larvae into two groups based on the ordination of the crotchets. Gardner (1948a) noted
differences in chaetotaxy, skin ornamentation, proleg number, maxillary palp and antennal
segment shape but made no subdivisions using these characters. A similar dichotomy in
chaetotaxy was found by Timlin (1955) but unfortunately, like Gardner, the number of species
examined was too small to be able to conclude anything definitive about the interrelationships of
the genera.

This position was changed markedly by Crumb’s (1956) study of the larvae of 25 species in 14
genera of Plusiinae. On the basis of characters drawn mainly from the mandibles and chaeto-
taxy, he grouped the plusiines into nine informal genera. Four of these (Abrostola, Polychrysia,
Chrysaspidia [Plusia] and Caloplusia) corresponded to McDunnough’s generic concepts.
Trichoplusia, Pseudoplusia and Argyrogramma were placed in a single taxon, as were Autoplu-
sia and Rachiplusia, and Plusia [Diachrysia] and Agrapha [Allagrapha]. Those species of
Syngrapha (rectangula (Kirby) and celsa (Edwards)) that possessed SV2 on segments TII and
TIII, and which also fed as larvae on conifers, were separated as ‘Genus no. 11’, while those
Autographa that possessed SV2 on segment Al (mappa (Grote & Robinson), ampla (Walker)
and corusca (Strecker)) were included in Syngrapha. Autographa was thus restricted to those
species lacking SV2 on segment Al — biloba (Stephens), californica (Speyer), gamma and
precationis (Guenée).

Beck (1960) examined the larvae of several European genera of Plusiinae: Abrostola, Plusia
[Diachrysia], Autographa, Chrysaspidia [Plusia], Syngrapha [Euchalcia] and Polychrysia, but
also drew no conclusions as to possible affinities.

Although these early larval studies made relatively few hard statements regarding the higher
classification of the Plusiinae, they did demonstrate that larvae could be of use taxonomically
and paved the way for future research.

Kostrowicki and the Palaearctic plusias

The next major revision of the Plusiinae was performed by Kostrowicki (1961), who studied the
Palaearctic fauna, comprising 83 species in 23 genera. Abrostola and Mouralia were omitted
from consideration. Kostrowicki was generally in favour of McDunnough’s (1944) arrangement

80 IAN J. KITCHING

of the Nearctic fauna but expressed reservations regarding its uncritical application to the
Holarctic fauna. Forbes’ (1954) reduction of McDunnough’s genera to subgeneric status was
rejected because this would ‘introduce quite unnecessary inequalities with the other subfamilies
of [Noctuidae]’. Crumb’s (1956) arrangement found more favour but Kostrowicki questioned
both the synonymy of Trichoplusia, Pseudoplusia and Argyrogramma, and the reality of
Crumb’s ‘genus no. 11’.

Kostrowicki then proposed his own phylogeny of the plusias, which he hoped would correct
some of McDunnough’s and Crumb’s parochial errors (Fig. 1). Argyrogramma was considered
to be the ancestral plusia lineage, which gave rise to three groups of genera. The first consisted of
Anadevidia, Pseudoplusia, Trichoplusia, Chrysodeixis, Macdunnoughia and Diachrysia [Thy-
sanoplusia]. This last genus, for some reason, was omitted from the phylogenetic tree. The
second group was characterized by a tendency for the palps to be rather longer than in the other
plusiines. One subgroup, Plusia [Diachrysia], Polychrysia and Panchrysia, was considered to be
possibly related to Diachrysia [Thysanoplusia]. The second subgroup consisted of a postulated

Anadevidia
Pseudoplusia
Trichoplusia
Macdunnoughia
Chrysodeixis
Argyrogramma
Diachrysia
Panchrysia
Polychrysia
Euchalcia
Plusidia
Lamprotes
Pseudeva
Plusia
Chrysanympha
Cornutiplusia
Autoplusia
Rachiplusia
Autographa
Anagrapha
Syngrapha

Caloplusia

Fig. 1 Phylogenetic tree of Kostrowicki. Originally, his tree depicted supposedly ancestral genera
(Argyrogramma, Diachrysia (as Plusia), Panchrysia, Euchalcia, Plusia (as Chrysaspidia) and Auto-
grapha) by thicker branches and branch lengths appear to have been related in some way to degree of
anagenetic change. These two aspects were eliminated and only the topology retained in the above
figure. Generic names have been changed into their modern forms. (Redrawn from Kostrowicki, 1961.)

SYSTEMATICS OF THE PLUSIINAE 81

ancestral genus, Euchalcia, and three derivative genera, Plusidia, Chrysoptera [Lamprotes] and
Pseudeva. Kostrowicki commented on Eosphoropteryx in connection with these genera but felt
he could not place it within the group because of the distinctive appearance of the genitalia.

The third group to be derived from Argyrogramma was based around Autographa and
Chrysaspidia [Plusia]. The latter was considered to have given rise to Chrysanympha, while the
former had given rise to two groups of genera, both characterized by spinose tibiae. The first
consisted of the Nearctic Autoplusia and Rachiplusia and the Old World genus Cornutiplusia.
The second included the primarily arctic-alpine genera Syngrapha and Caloplusia, and the more
widely spread Nearctic Anagrapha, of which Syngrapha was considered to be the ancestral
group. Kostrowicki rejected Crumb’s (1956) delimitations of Autographa and Syngrapha
because the divisions based upon the larvae were not reflected by differences in the genitalia.

Kostrowicki’s scheme showed several important discrepancies compared to the generic
interrelationships that had been proposed previously. Several were due to incorrect assignment
of species to genera, which then became structurally heterogeneous, or to straightforward
misidentification (Ichinosé, 1963a; Dufay, 1970a). Into the first category fell, for example, the
placing of Autographa aemula ({Denis & Schiffermiiller]), A. bractea and A. argyrosigna
(Moore) in Chrysaspidia [Plusia]. The inclusion of Antoculeora ornatissima and Erythroplusia
rutulifrons in Argyrogramma, and Erythroplusia pyropia (Butler) in Anadevidia were due to
misidentifications. The male genitalia of Antoculeora ornatissima figured have the quadrate
saccus typical of Argyrogramma and are probably those of Argyrogramma signata (Fabricius)
(Dufay, 1970a: 150, footnote), the only species of Argyrogramma (s. str.) to occur in eastern
Asia. ‘Argyrogramma rutulifrons’ appears to be very closely related to Argyrogramma circums-
cripta (Freyer), and may even be that species. The identity of ‘Anadevidia pyropia’ remains a
mystery.

The rising influence of larvae: Ichinosé’s revisions

Following on from a series of studies of plusiine larval morphology (Ichinosé, 1958a; b; 1960a;
1962a), Ichinosé (1962b) produced a revision of the Japanese Plusiinae, contemporaneous with
Kostrowicki’s work. Ichinosé considered that McDunnough had oversplit the subfamily and that
many of the characters employed were of little taxonomic significance, particularly the form of
the clavus. Instead; Ichinosé preferred the generic assignments suggested by Crumb (1956),
although he felt that Crumb had placed too much emphasis on certain characters and thus had
reached some dubious conclusions. Although Ichinosé was in favour of using characters drawn
from all stages of the life-history in the construction of classifications, at the generic level he
predominantly used larval characters, particularly from the chaetotaxy.

Ichinosé recognized twelve genera of Plusiinae. The delimitation of eight of these was
uncontroversial and conventional: Abrostola, Plusidia, Chrysoptera [Lamprotes], Euchalcia,
Podioplusia [Anadevidia], Chrysaspidia [Plusia] and Syngrapha. Autographa was divided into
three subgenera: Autographa (s.str.), Scleroplusia [Macdunnoughia (s.1.)| and Erythroplusia,
separated on the basis of the divergent form of the male genitalia but united by a uniform larval
morphology. Ichinosé, like Kostrowicki, separated intermixta from the remaining species of
Plusia [Diachrysia] and placed it, together with daubei (Boisduval) and ochreata, in Diachrysia
[Thysanoplusia]. The golden wing pattern of intermixta and the genus Plusia [Diachrysia] was
considered to be convergent.

While this reallocation of species left Plusia [Diachrysia] more compact and homogeneous,
the three species of Diachrysia [Thysanoplusia] showed diverse possible affinities. Overall,
Ichinosé considered Diachrysia [Thysanoplusia] to be much more closely related to Chry-
sodeixis than to Plusia [Diachrysia]. However, while the chaetotaxy and skin ornamentation of
intermixta suggested further affinities with Podioplusia [Anadevidia] and Euchalcia, the chaeto-
taxy and external appearance of daubei suggested a possible connection with Chrysodeixis.

With regard to Chrysodeixis itself, Ichinosé followed Crumb (1956), including Trichoplusia
(and also the species now assigned to the nominal genera Ctenoplusia (albostriata) and
Acanthoplusia (agnata (Staudinger) and farassota)).

82 IAN J. KITCHING

Ichinosé presented his overall view of the generic relationships in the form of a ‘phylogenetic
tree’ (Fig. 2). Exactly how this dendrogram is to be interpreted is uncertain. It agrees reasonably
well with Kostrowicki’s tree (Fig. 1), in that the Polychrysia-group and the Autographa-group
arose independently from a Chrysodeixis-Argyrogramma-type ancestral form. In both treat-
ments, Podioplusia [Anadevidia] was of uncertain position, while Ichinosé’s clarification of the
positions of Macdunnoughia and Erythroplusia represented a considerable improvement over
Kostrowicki’s placings.

Meanwhile, Claude Dufay had been busily publishing a large number of studies on plusiine.
alpha-taxonomy. He eventually ventured his opinions on the higher classification of the
subfamily in his revision of the Madagascan fauna (Dufay, 1970a). Following the obligatory
discussion of Hampsonian type species designations, Dufay proceeded with a brief discussion of
the studies of McDunnough, Kostrowicki and Ichinosé. Like Ichinosé (1963), Dufay disagreed
with some of Kostrowicki’s generic assignments, particularly regarding Argyrogramma. How-
ever, he also considered Ichinosé’s treatment to be unsatisfactory, for although the groupings
were certainly more natural than those of Kostrowicki, several genera (in particular Chryso-
deixis) were too heterogeneous. Dufay preferred to keep Trichoplusia, Argyrogramma and his
own Ctenoplusia separate from Chrysodeixis.

Like Kostrowicki, Dufay’s revision was also reviewed in Japan (Sugi, 1972). The major
criticism appeared to be Dufay’s inclusion of /Thysanoplusia] in Trichoplusia, which thereby
rendered that genus heterogeneous in terms of larval chaetotaxy and skin ornamentation.

Shortly afterwards, Ichinosé (1973) published an upgraded classification of the Japanese
plusiines. Autographa now included Macdunnoughia and Sclerogenia as subgenera but the
previously included subgenus Erythroplusia was elevated to full generic status. Erythroplusia
comprised two subgenera, Erythroplusia (s.str.) and Antoculeora, united by characters of the

® Abrostola
Fi

Y @ Chrysodeixis
Ctenoplusia
Diachrysia
Thysanoplusia
Anadevidia
Polychrysia
Euchalcia
Plusidia
Lamprotes
Erythroplusia
Macdunnoughia
Autographa

Plusia

Syngrapha

Fig. 2. Phylogenetic tree of Ichinosé. This tree originally consisted of a series of interconnected circles
representing the genera. Thus several taxa (Abrostola, Chrysodeixis, Ctenoplusia (as Chrysodeixis)
agnata, Euchalcia, Thysanoplusia (as Diachrysia), Erythroplusia and Autographa (s.str.)) were inter-
preted as ancestral to other genera. These have been placed in the above figure on the ends of
‘zero-length’ branches, represented by dashed lines. Only two character complexes were included on
Ichinosé’s tree; larval chaetotaxy type and the presence/absence of lateral tufts on the adult male
abdomen. Generic names have been changed to their modern forms. (Redrawn from Ichinosé, 19625).

SYSTEMATICS OF THE PLUSIINAE 83

asymmetrical male genitalia, larval morphology and, tentatively, larval foodplant. However,
some possible affinity between Antoculeora and Macdunnoughia was admitted on the basis of
the female genitalia (the sole species of Antoculeora, ornatissima, had been previously included
by Ichinosé [19626] in Macdunnoughia). Ichinosé (1973) remained committed to the view that
the Chrysodeixis-Trichoplusia-Argyrogramma complex formed a homogeneous group. To
recognize a multitude of genera, as had Dufay (1970a), was regarded as ‘myopic splitting’.
However, Ichinosé did relent somewhat and accepted Chrysodeixis (with Pseudoplusia as a
tentative synonym) as distinct from Argyrogramma. This latter genus was divided into three
subgenera: Argyrogramma, Trichoplusia and Ctenoplusia. Agrapha [Allagrapha] was synony-
mized again with Plusia [Diachrysia] because the two shared many similarities in the form of
abdominal segment 8 in the adult female and in the larval chaetotaxy (but see Eichlin &
Cunningham, 1978, and below).

At this point, I will digress briefly from the history of the higher classification of the Plusiinae
to examine the problems associated with the type species of the genera Plusia and Diachrysia,
and the application of these names. Up until 1970, most authors had treated orichalcea sensu
Hiibner, designated by Dyar (1902), as the type species of Diachrysia. This proved to be a
misidentification of Diachrysia chryson and resulted in the orichalcea-intermixta group of
species, which were certainly not congeneric with chryson (Kostrowicki, 1961; Ichinosé, 19625),
being bereft of a genus. Ichinosé (1973) resolved this by proposing Thysanoplusia.

With regard to Diachrysia and Plusia, most authors had treated these genera as synonyms. In
this they followed Grote (1895) in accepting Duponchel’s (1829) designation of chrysitis as the
type species of Plusia, a species that was then, and still is, considered to be congeneric with
Diachrysia chryson. However, they had all overlooked Duponchel’s (1826) previous designa-
tion of festucae as the type species. Thus Plusia correctly refers to the festucae-group of plusiines
and Chrysaspidia sinks as a synonym. The next available name for the chrysitis-chryson group is
Diachrysia. However, as noted above (and by Dufay, 1970a, and Nye, 1975), this name involves
a case of misidentified type species, which should therefore be referred to the International
Commission on Zoological Nomenclature under Article 70(b) of the Code. Although this will be
done in the future, the consensus of all modern works (e.g. Franclemont & Todd, 1983;
Bretherton ef al., 1984; Skinner, 1984) is that chryson is the type species of Diachrysia.
However, the erroneous inclusion of orichalcea in Diachrysia is still prevalent (e.g. Bretherton
et al., 1984; Skinner, 1984).

To return to the higher classification of the Plusiinae, a totally independent test of all previous
arrangements was performed by Nakamura (1974), using characters derived from the hitherto
generally ignored pupal stage. Mosher (1916) and Gardner (1948b) had examined a small
number of species but Nakamura studied 21 species, which allowed him to comment on possible
generic interrelationships. Very few of the groupings that he found corresponded to those
previously proposed and several cut across the generic boundaries of Kostrowicki, Dufay and
Ichinosé. Congeners in Chrysodeixis (s.l. or s.str.), Thysanoplusia and Acanthoplusia showed
marked differences in pupal chaetotaxy, while, in contrast, species of the genera Abrostola,
Polychrysia, Plusia [Diachrysia] and Autographa (s.str., sensu Ichinosé, 1962b) had discrete
pupal types, in terms of structure and chaetotaxy. Nakamura compared his results with the
generic concepts of Ichinosé (19626), Dufay (1970a) and Esaki et al. (1971) but refrained from
making any formal taxonomic changes of his own.

However, the latest Japanese checklist (Sugi, 1982) has returned to an almost maximally split
arrangement. Lamprotes, Macdunnoughia, Antoculeora, Sclerogenia, Trichoplusia. Ctenoplu-
sia and Acanthoplusia were all treated as distinct genera. Only Thysanoplusia (N.B.) and
Puriplusia were recognized as subjective synonyms, being sunk to Trichoplusia and Macdun-
noughia respectively.

The focus moves back west: Eichlin & Cunningham’s reappraisal of the Nearctic fauna

Meanwhile, in the U.S.A., analysis of the American Plusiinae had proceeded along similar lines
as had the Japanese fauna a decade earlier. A series of studies of larvae (Eichlin & Cunningham,

84 IAN J. KITCHING

1969; Eichlin, 1975) led eventually to a full scale revision of the subfamily on a regional basis
(Eichlin & Cunningham, 1978). Like Ichinosé, Eichlin & Cunningham employed both larval
and adult characters in the construction of their classification. However, they tried not to place
undue emphasis on either data source and so hoped to arrive at an arrangement that was less
conservative than that of Ichinosé but not as split as those of McDunnough, Kostrowicki or
Dufay. The end result, however, did tend more towards the classifications of the latter three
authors (Fig. 3).

Largely on the basis of the form of the larval prolegs, Eichlin & Cunningham divided the
subfamily into three tribes, the first occasion on which this taxonomic rank had been used in the
Plusiinae. The most primitive tribe, in which the prolegs on A3 and A4 (see Terminology) were
fully functional, if albeit slightly reduced, was the Abrostolini, comprising Abrostola and
Mouralia. The latter was seen as a derivative of the former, having more specialized genitalia.

Although Eichlin & Cunningham appeared to treat the Abrostolini as a monophyletic unit,
they considered that ‘a common ancestor from the tinctoides [as opposed to the Abrostolini] line
gave rise to the Argyrogrammini [Argyrogrammatini] and Autographini [Plusiini]’. In both of
these tribes, the ribs on the larval mandibles bore subterminal processes and the prolegs on A3
and A4 were non-functional.

The Argyrogrammini [Argyrogrammatini] were interpreted as the more primitive tribe
because its members retained vestigial prolegs whereas all traces of prolegs had been lost in the
Autographini [Plusiini]. Eichlin & Cunningham retained the three genera of McDunnough
Abrostola
Mouralia
Argyrogramma
Trichoplusia
Pseudoplusia
Autoplusia
Rachiplusia
Diachrysia
Allagrapha
Polychrysia
Adeva
Pseudeva
Chrysanympha

Eosphoropteryx
Autographa

Syngrapha

Plusia

Fig. 3 Generic phylogeny of the Plusiinae according to Eichlin & Cunningham. Their original dendro-
gram included an anagenetic component that is omitted in the above figure. The positions of Eosphoro-
pteryx and Chrysanympha were uncertain, as indicated by dashed lines. Autographa was interpreted as
being biphyletic. The detailed interrelationships of the species of Autographa and Syngrapha are
omitted. (Redrawn from Eichlin & Cunningham, 1978.)

SYSTEMATICS OF THE PLUSIINAE 85

(Argyrogramma, Trichoplusia and Pseudoplusia), rejecting Ichinosé’s (1973) synonymy of the
first two, and of the third with Chrysodeixis, on the grounds that Ichinosé’s larval characters
broke down when applied to the Nearctic species. Eichlin & Cunningham preferred to steer a
middle course between Ichinosé’s larval dominated generic concepts and Dufay’s adult domin-
ated system, recognizing, in addition, that the results of Nakamura’s (1974) pupal study were at
odds with both. However, there were too few Nearctic Argyrogrammini [Argyrogrammatini] to
permit Eichlin & Cunningham to draw any significant conclusions regarding the generic limits
and interrelationships within the tribe. The three genera were represented on the phylogenetic
tree (Fig. 3) as a trichotomy, although Pseudoplusia was considered to be the most advanced
genus because it had lost the lateral abdominal tufts in the male (present in Trichoplusia and
Argyrogramma), retained the mandibular processes (lost in the other two genera) and had a
unique highly-elongate apex to the corpus bursae.

The third tribe of plusiines, the Autographini [Plusiini], contained the balance of the
American genera, 12 in all. Eichlin & Cunningham refrained from subdividing the group using
adult characters because of the high degree of larval uniformity. The first group of four genera to
branch off the main phylogenetic line were characterized by the retention of separate pinacula
for setae SV1 and SV2 on A2. In all other genera of Autographini [Plusiini] examined, these two
pinacula were at least partially fused. Using the degree of separation of the V1 setae on Al—4
(supposedly linked to the time elapsed since the prolegs were lost on A3—4), Autoplusia was
interpreted as the most primitive genus, followed by Rachiplusia, Plusia [Diachrysia] and
Allagrapha. However, this arrangement was complicated by the presence of tibial spines, long
believed to be an advanced character, in the former two genera.

With regard to Allagrapha, Eichlin & Cunningham rejected its synonymy with Plusia
[Diachrysia]. In particular, the presence of a raduloid (a ridged structure on the larval
hypopharynx) in the two species of Allagrapha (absent in the Nearctic Plusia [Diachrysia]
balluca Geyer, the only species of the genus examined as a larva), features of the male genitalia
and a strong superficial similarity were used to argue full generic status for Allagrapha.

The next section of Eichlin & Cunningham’s tree was highly tentative because the larvae of
four of the five genera (Adeva, Pseudeva, Eosphoropteryx and Chrysanympha) were not known.
The fifth, Polychrysia, was considered to form the next branch off the main stem, because the
larva retained the primitive arrangement of the V setae mentioned above. Adeva and Pseudeva
were placed together because of the very reduced clavus and associated with Polychrysia
because of the overall similarity of their genitalia. Chrysanympha and Eosphoropteryx were
inserted next on the tree but their exact relationships were left open.

Eichlin & Cunningham noted that Autographa, while structurally uniform in the male and
female genitalia, could be subdivided into two lineages by the presence or absence of seta SV2
on A1, presence being the derived condition. The final two genera, Syngrapha and Chrysaspidia
[Plusia], were well defined. The former had spinose tibiae and the latter had uniordinal
crotchets (unique in the Plusiinae) and an unusual larval mandible. Syngrapha was further
subdivided into a number of species groups, one of which (including Palaeographa and
Caloplusia) was characterized by the presence of seta SV2 on TII and TIII, a feature unique
within the Noctuidae (see also Crumb, 1956).

Eichlin & Cunningham concluded by noting that the three major classifications, their own and
those of Kostrowicki and Ichinosé, were not particularly different in terms of the major
evolutionary lineages they depicted. They considered this to be highly significant, given the
more or less exclusive nature of the faunas upon which each was based. Nevertheless, Eichlin &
Cunningham felt that there was still a great gap in the knowledge of the tropical plusiines and
that this needed to be filled before a synoptic classification of the subfamily could be established.

The latest North American checklist (Franclemont & Todd, 1983) explicitly followed Eichlin
& Cunningham’s tribal and generic assignments but within a few (unexplained) changes.
Trichoplusia oxygramma was removed into Agrapha, under which was synonymized Ctenoplu-
sia and Acanthoplusia. Allagrapha aereoides Grote (as aeroides) was put back into Diachrysia
(following McDunnough, 1944) and Adeva was synonymized with Euchalcia (following Kostro-
wicki, 1961). Anagrapha was resurrected for the divergent falcifera.

86 IAN J. KITCHING
Return to the east: Chou & Lu’s phylogeny

The late 1970s saw a surge of interest in the Plusiinae of China. Following a series of descriptive
papers (Chou & Lu, 1974; 1978; 1979a; b), which included many inadvertent synonyms due
probably to poor access to previous publications on the subfamily, Chou & Lu (1979c) published
a summary of their ideas on the higher classification of the Plusiinae (Fig. 4).

Omorphina
Abrostola
Euchalcia
Lamprotes
Plusidia
Polychrysia
Panchrysia
Anadevidia
Thysanoplusia
Diachrysia
Caloplusia
Syngrapha
[Neoplusia ]
Shensiplusia
Chrysodeixis
Trichoplusia
Dactyioplusia
Zonoplusia
Argyrogramma
Acanthoplusia
Puriplusia
Erythroplusia
Plusia
Cornutiplusia
Autographa
[Scleroplusia]
Macdunnoughia

Antoculeora

Fig. 4 Phylogeny of the Plusiinae according to Chou & Lu. Seven tribes were recognized (Omorphinini,
Abrostolini, Euchalciini, Plusiini [sic], Caloplusiini, Argyrogrammini [sic] and Autographini) corres-
ponding to the seven major lineages. Neoplusia and Scleroplusia are synonyms of Chrysodeixis and
Macdunnoughia respectively and are placed in square brackets. Generic names have been converted
into their modern forms. (Redrawn from Chou & Lu, 19795.)

SYSTEMATICS OF THE PLUSIINAE 87

Seven tribes were recognized. The most primitive was considered to be the Omorphinini,
comprising only Omorphina aurantiaca. Chou & Lu acknowledged the earlier suggestion
(Kostrowicki, 1961) that Omorphina ought probably be transferred to the Erastriinae [Acon-
tiinae], but they chose to retain it in the Plusiinae pending more detailed study. The next tribe,
the Abrostolini, contained only Abrostola. Euchalcia, Chrysoptera [Lamprotes], Plusidia,
Polychrysia and Panchrysia were placed in the Euchalciini, while the [Plusiini] comprised Plusia
[Diachrysia], Thysanoplusia and Anadevidia. The Caloplusiini included Syngrapha and Calo-
plusia. By far the largest tribe was the Argyrogrammini [Argyrogrammatini], consisting of four
subgroups; two contained only single genera (Trichoplusia, Dactyloplusia), a third included
Neoplusia [Chrysodeixis], Shensiplusia and Chrysodeixis, while the fourth consisted of a
pentachotomy subtending Zonoplusia, Argyrogramma, Acanthoplusia, Puriplusia and Erythro-
plusia. These last two genera were placed near Acanthoplusia despite lacking vestigial larval
prolegs (also absent in A. agnata; Ichinosé, 1962b) and anal tufts in the male. If Puriplusia and
Erythroplusia are excluded, then the Argyrogrammini [Argyrogrammatini] is more or less
equivalent to Chrysodeixis (s.1.) (sensu Ichinosé, 1973).

The final tribe, the Autographini [Plusiini], included six genera in three subgroups. Chrysaspi-
dia [Plusia] represented the most basal branch, followed by Autocureola [sic, Antoculeora]. The
four remaining genera (Autographa, Cornutiplusia, Scleroplusia [Macdunnoughia] and Mac-
dunnoughia) were treated as a tetrachotomy. Cornutiplusia was considered to be identical to
Autographa (except for the spined tibiae) and Autocureola [Antoculeora] was interpreted as the
most advanced genus.

No resolution of the tribal interrelationships was attempted. Several of the tribes, notably the
[Plusiini], were distinctly phenetic in construction. Chou & Lu preferred to group genera ona
limited number of superficial characters of doubtful taxonomic significance (see below), which
resulted in a number of generic misplacements. Nevertheless, the overall classification had a
great deal in common with those proposed previously.

Ronkay (1982), in a review of the Korean fauna, also accepted the general findings of
Kostrowicki, Ichinosé and Eichlin & Cunningham. However, he disagreed on several points.
Ronkay considered Ichinosé’s interpretation of Macdunnoughia as a subgenus of Autographa
was incorrect and he reinstated the former to generic level. Chou & Lu’s (1974; 1979c) treatment
of Puriplusia as a good genus was also disputed and Ronkay, agreeing with Dufay (1977),
reduced Puriplusia to subgeneric rank within Macdunnoughia. The generic assignments within
the Argyrogrammatini were also considered problematical and, pending a full tribal revision,
Ronkay chose to retain the genera Trichoplusia, Ctenoplusia, Chrysodeixis and Argyrogramma
according to Dufay’s usage.

Holloway (1985), in his study of the Bornean fauna, also maintained Chrysodeixis and
Ctenoplusia as distinct, pending further investigation. Thysanoplusia, Zonoplusia and Dactylo-
plusia were also treated as good genera but several plusiine species remained unassignable to
current genera. Holloway followed the practice of Dufay (e.g. 1972; 1974) in placing such
species in ‘Plusia’ s.1. Although this approach is unsatisfactory, it was the best that could be
achieved for these species until a comprehensive revision of the subfamily, and particularly the
Argyrogrammatini, is performed.

Kljutshko’s ‘cladistic’ analysis of the Plusiinae

The most recent studies to consider the higher classification of the Plusiinae are those of
Kljutshko (1985a; b). For the first time in plusiine systematics, cladistic techniques were overtly
employed to analyse a data set of 32 characters, drawn from larval, pupal and adult morphology.
The result was a cladogram (Fig. 5) based upon the 24 characters that Kljutshko considered to be
the most well-founded. However, the analysis was fundamentally flawed in such ways as to
render most of the resolution of the cladogram spurious and thus invalidate most of the
conclusions drawn from it.

First, and most crucial, the cladistic methodology was misapplied. Although some groups
were supported by putative synapomorphies (e.g. Ctenoplusia and Acanthoplusia on the basis of

88 IAN J. KITCHING
Autographa

Desertoplusia
Diachrysia
Macdunnoughia
Antoculeora
Plusia
Plusidia
Euchalcia
Pseudochalcia
Anadevidia
Lamprotes
Panchrysia
Polychrysia
Cornutiplusia
Syngrapha
Caloplusia
Trichoplusia
Argyrogramma
Chrysodeixis
Zonoplusia
Ctenoplusia
Acanthoplusia
Erythroplusia

Abrostola
Fig.5 Cladogram of the plusiine genera according to Kljutshko. (Redrawn from Kljutshko, 1985a, b.)

blade-like setae on the valve), others were distinguished and delimited using symplesiomor-
phies. For example, Autographa and Desertoplusia were united, and separated from Diachrysia
on the basis of a long harpe, the plesiomorphic state for his character 22. Similarly, the members
of the ‘clade’ Autographa-Antoculeora were distinguished from Plusia by the presence of
biordinal crotchets on the larval prolegs (Kljutshko’s character 6). However, all other known
plusiine larvae have biordinal crotchets and it is more parsimonious (and in keeping with
previous authors’ conclusions) to treat the uniordinal condition in Plusia as autapomorphic for
this genus. This would appear to be a failure to employ the outgroup criterion correctly.

If character 6 is reinterpreted and all the groups characterized by symplesiomorphies
collapsed to form polytomies, then a much less well-resolved cladogram results (Fig. 6). Three
tribes were recognized by Kljutshko: the Abrostolini, the Argyrogrammatini (including
Erythroplusia) and the Plusiini. The last consisted of three subtribes: the Plusiina (including
Anadevidia), the Caloplusiina and the Polychrysiina. Of the tribes, the Abrostolini and the
Argyrogrammatini were sister-groups, in contrast to every other phylogeny that had been
previously proposed for the Plusiinae. This relationship was based upon the possession of a

SYSTEMATICS OF THE PLUSIINAE 89
Autographa

Desertoplusia
Diachrysia
Macdunnoughia
Antoculeora
Plusia
Plusidia
Euchalcia
Pseudochalcia
Anadevidia
Lamprotes
Panchrysia
Polychrysia
Cornutiplusia
Syngrapha
Caloplusia
Trichoplusia
Argyrogramma
Chrysodeixis
Zonoplusia
Ctenoplusia
Acanthoplusia
Erythroplusia

Abrostola

Fig. 6 Cladogram of the Plusiinae derived from the data set of Kljutshko (19855) following the correct
application of cladistic methodology. Many of the taxa in Fig. 5 were delimited by plesiomorphic
character states. Their elimination results in numerous polytomies.

finer, hair-like seta SD1 on A9 of the larvae (this state is also convergently derived in
Cornutiplusia).

However, even the reduced cladogram shown in Fig. 6 is defective. The reason for this lies in
the characters that Kljutshko used, many of which are insubstantial and nugatory (e.g. the forms
of the clavus and harpe, and the origin of the ductus bursae on the corpus bursae), and all of
which appear to have been culled from the works of other authors. The result suffers from
deficiencies similar to those in the ‘phylogenetic tree’ of Chou & Lu (1979c).

Within the constraints imposed by Kljutshko’s choice of characters, the following additional
modifications can be made to the character analysis (albeit in the light of my own analysis
described below). In addition to character 6, the polarity of character 26 should be reversed
because the absence of a signum is plesiomorphic for the Plusiinae as a whole. Since the absence
of an epicranial suture in the pupa characterizes the members of both the Plusiini and the
Argyrogrammatini (see also Nakamura, 1974), this state can be reinterpreted as synapomorphic

90 IAN J. KITCHING

for these two tribes. It also follows from this that the polarity of character 4 (SD1 on larval A9)
should be reversed as well. I also reject characters 12 (only Euchalcia gerda and E. serraticornis
Dufay [Dufay, 1968] have serrate antennae, and only Lophoplusia has bipectinate antennae,
none of which were examined in Kljutshko’s study), 18 (if the dorsal abdominal tuft on the adult
A2 is reduced in any genus, it is in Plusia, not Syngrapha; see almost any author on the Plusiinae)
and 29 (the position of the ostium bursae on A8) as having any significance. The final result of
these changes is shown in Fig. 7 and represents the only informative structure present in
Kljutshko’s original cladogram (Fig. 5).

As alluded to above, the Argyrogrammatini and Plusiini have a sister-group relationship but a
trichotomous one that includes Erythroplusia as an unplaced genus. The members of the
Argyrogrammatini are united by the presence of subdorsal tufts on AS—6 in the male but form an

Autographa
Desertoplusia
Diachrysia
Macdunnoughia
Antoculeora
Plusia
Plusidia
Euchalcia
Pseudochalcia
Anadevidia
Lamprotes
Panchrysia
Polychrysia
Cornutiplusia
Syngrapha
Caloplusia
Trichoplusia
Argyrogramma
Chrysodeixis
Zonoplusia
Ctenoplusia
Acanthoplusia
Erythroplusia

Abrostola

Fig. 7 Cladogram of the Plusiinae derived from the data set of Kljutshko (1985b) following further
character analysis. Several characters (see text) were reinterpreted by the correct application of the
outgroup criterion for determining character polarity and several more were omitted. The resulting
cladogram represents the sum total of the information in Kljutshko’s data set. There are extensive
polytomies and few informative groups. Note, however, the monophyly of the Argyrogrammitini,
Ctenoplusia (s.1.) and the clade Macdunnoughia + Antoculeora.

SYSTEMATICS OF THE PLUSIINAE 91

apical pentachotomy, in which the only further resolution is the union of Ctenoplusia and
Acanthoplusia on the basis of blade-like setae on the valve. The Plusiini are characterized by a
normal seta SD1 on A9 and the total loss of prolegs in the larvae. The 16 constituent genera of
this tribe also form a highly polytomous apex, with no fewer than ten branches. Euchalcia and
Pseudochalcia are united by the form of the juxta, while Macdunnoughia and Antoculeora share
a weak uncus that is not apically hooked. Polychrysia, Lamprotes and Panchrysia are placed
together by the possession of very long palps and probosces but form a trichotomy. Cornutiplu-
sia, Syngrapha and Caloplusia are united by the shared presence of tibial spines, while the latter
two genera are treated as sister-groups because they both display a juxta with an apical process
(the same state that unites Euchalcia and Pseudochalcia). All the genera have autapomorphies
that will not be considered further here, with the exception of Chrysodeixis, Ctenoplusia,
Autographa, Polychrysia, Lamprotes and Syngrapha.

I have conducted the preceding re-analysis of Kljutshko’s cladogram in order, not to deride
his work, but to congratulate him. He, amongst all previous authors (with the possible exception
of Ichinosé, 1962b), has produced a thoroughly explicit analysis that is open to examination and
constructive criticism. Many other authors (notably Kostrowicki [1961] and Chou & Lu [1979c])
chose not to state the exact reasons (if any) for the conclusions they reached. Even Eichlin &
Cunningham (1978) only explicitly stated some of the characters they used in order to reach their
final classification, which rendered many parts of their ‘phylogenetic tree’ immune to criticism.
Thus, despite the errors in analytical procedure and his reliance upon a small number of largely
superficial characters (see below), Kljutshko did produce an arrangement for which his
reasoning was clear and in which the characters used to delimit each taxon were unambiguously
stated.

Two unconventional approaches -

All the studies discussed above investigated the external morphology of the larvae, pupae and
adults, with primary attention being focussed on the cuticular structures of the male and female
genitalia. There have been two further studies that considered more unconventional organ
systems. Both deal with only a very restricted sample of species but the results do appear to
corroborate some aspects of previous classifications. For this reason, they will be discussed in
detail.

Callahan (1960) examined the structure of the male and female reproductive tracts in eight
species of Plusiinae, paying particular attention to those parts that operate in the formation and
insertion of the spermatophore. Thus, in addition to the aedeagus and the corpus bursae,
Callahan also examined the cuticular simplex (the part of the male genitalia between the testes
and the aedeagus that is normally removed in genitalia preparations) and the spermatophore
itself.

The species were found to fall into two distinct groups. Group I included Autographa biloba,
Agrapha [Allagrapha] aerea and Rachiplusia ou (Guenée), and thus contained only and all the
members of the Plusiini studied. They were characterized by a spermatophore with a short,
heavy collum and large cap, a short aedeagus without a sclerotized bulbous lock and a cuticular
simplex with a short, broad constrictor muscle. In contrast, the species of Group II had a
medium or long, narrow collum, a small spermatophore cap that fits tightly at the apex of the
corpus bursae where the ductus seminalis arises, a sclerotized bulbous lock on the inner wall of
the uneverted vesica, a long aedeagus and a constrictor muscle that is elongate and tapered at the
basal end. This group included Pseudoplusia includens, Trichoplusia ni, T. oxygramma (Geyer),
Argyrogramma verruca and A. basigera (Walker), and thus corresponded to the Argyrogram-
matini. Although there were characters that might have allowed subdivision of these groups
(particularly states that might group Autographa and Agrapha [Allagrapha] to the exclusion of
Rachiplusia; and Trichoplusia and Pseudoplusia to the exclusion of Argyrogramma), such
analysis was precluded by the lack of an outgroup by which to polarize the transformation series.
However, the results of Callahan’s study did corroborate the tribes subsequently proposed by
Eichlin & Cunningham (1978).

92 IAN J. KITCHING

The second ‘unconventional’ morphological study examined the musculature of the male
genitalia (Bayer, 1965). The results of a previous (unreferenced) study of the valves of 25 South
African plusiine species suggested that they could be divided into five groups. Four (A-D)
corresponded to Cornutiplusia, Chrysodeixis, Ctenoplusia and Trichoplusia (part). The fifth
group, E, included two species of Thysanoplusia, two species of Trichoplusia (sestertia (Felder)
and arachnoides (Distant)) and ‘Plusia’ tetrastigma (Hampson) (which may properly belong in
Trichoplusia [Dufay, note in BMNH collection]). Finally, there were four unplaced species: two
Abrostola and two Trichoplusia (transfixa (Walker) [vittata (Wallengren)] and obtusisigna
(Walker)). The distinction between groups D and E was noted as not being very marked. Bayer
then proceeded to examine the male genital musculature of six species in greater detail:
Cornutiplusia circumflexa, Chrysodeixis acuta (Walker), Ctenoplusia limbirena, Trichoplusia
angulum (Guenée), T. transfixa [vittata] and Thysanoplusia orichalcea. (The first species was
placed in Syngrapha, the remainder in Plusia.)

Although a number of structural differences were found to suggest Cornutiplusia, Chry-
sodeixis and Trichoplusia vittata might be separated from the other three species, no characters
were found that would enable their grouping into higher taxa. Nevertheless, the discovery of
quite major musculature changes, such as the loss of entire muscles or marked differences in
insertions, does suggest that the genital musculature has great potential in higher classification.

The position of the Plusiinae within the Noctuidae

When the Plusiinae were first proposed as a group (Boisduval, 1828), they were placed as part of
a linear sequence between the Noctuelidi, a conglomerate of the presently recognized trifine
subfamilies, and the Heliothidi. Guenée (1841) and Duponchel (1844) retained a tentative
relationship with the Heliothidi, but also favoured an affinity with the Calpidi. In contrast,
Herrich-Schaffer (1845) removed the Plusidae [sic] from any association with the Heliothidae,
and placed them in a part of his sequence that included the Metoponidae and Erastridae.

Guenée’s second classification (1852-4) divided the noctuids into two main phalanges based
on hindwing venation. The Plusidae were included in the tribe Variegatae of the phalange
Quadrifidae, together with such families as the Eurhipidae [Euteliinae], Calpidae and Gonop-
teridae. Thus, in his revised classification, Guenée retained a relationship between the Plusidae
and the Calpidae, but rejected a connection with the Heliothidae, which was placed in the
phalange Trifidae.

Packard (1869) also employed a two-way split of the Noctuidae but based upon the shape of
the clypeus and the antennal form, rather than wing venation and palps. Thus, he included the
plusiines in the Noctuinae with the trifine noctuids, rather than in the Catocalinae, with the
quadrifines. Grote (1882; 1895) continued this usage, which reassociated the Plusiinae with the
Heliothinae, but also placed the Calpinae nearby, together with the Stiriinae, which were
considered to be closely related to the heliothines.

Both Tutt (1891-2) and Hampson (1893-5) preferred to use versions of Guenée’s (1852-4)
second classification. Tutt (1902) subsequently modified his position to approximate that of
Grote (1895) by considering the plusiines to be a highly developed group of heliothines. The
reduced larval prolegs of most plusiines, by which they were associated with the Calpinae and
other quadrifines with semi-looper larvae, were explained as the result of convergent evolution.

In contrast, Hampson (1902) had completely reappraised his classification of the Noctuidae
and this culminated in a system that was at once to revolutionize and stultify noctuid higher
systematics. By classifying the noctuid moths into semi-artificial assemblages, using a small
number of characters applied absolutely, Hampson succeeded in creating a series of subfamilies
between which there were no possibilities for higher level relationships. Noctuid higher
classification, which had been hierarchical to varying degrees since Boisduval (1828), had been
reduced to a single tier system in one fell swoop. The Plusiinae were no longer seen as having
closer relationships with some subfamilies than others; now they were just one among 15 equals.
Even the major division between trifine and quadrifine hindwing venation was not reflected in
the classification.

SYSTEMATICS OF THE PLUSITINAE 93

This attitude towards higher level classification was further institutionalized by the authors of
‘Seitz’, but this unfortunate state of affairs did not long go unchallenged. Richards (1932)
conducted a detailed comparison of the morphology and evolution of the tympanal organ in
which, in addition to dividing the Noctuidae into a number of subfamilies, he also suggested
possible interrelationships between the subfamilies in the form of a phylogenetic tree. The
Plusiinae were found to exhibit a very distinct type of tympanal organ, with a double tympanal
hood and a greatly enlarged pocket IV, intimately fused to the tympanal air sac. The subfamily
Stictopterinae were found to possess a tympanal hood with a ventral lobe and a pocket IV which,
while of normal size for a noctuid, was intimately fused with the air sac as in the Plusiinae. On the
basis of these similarities, Richards proposed that the Plusiinae had been derived via the
Stictopterinae from the higher Erebinae [Catocalinae] of group IV (including Euparthenos
Grote and Zale Hubner). This relationship between the Stictopterinae and the Plusiinae was
accepted by all subsequent authors, Forbes (1954) even going so far as to suggest that the two
subfamilies might better be treated as one. Eichlin & Cunningham (1978) also found no fault
with this suggestion and used it in order to construct their ‘prototype’ plusiine. One of the major
conclusions of the work detailed below is that the belief in such a stictopterine-plusiine
sister-group relationship is erroneous.

Taxa examined

The 400 or so species of Plusiinae are currently classified into about 30 to 50 genera, depending
upon the degree of taxonomic ‘splitting’ adopted. To attempt to maximise the resolution of the
supraspecific relationships and to test the validity of many of the synonymies, 45 nominal genera
were selected for analysis, all of which have been in recent use. The type species of each
(Appendix 1) was employed as an exemplar taxon and although difficulties were encountered in
coding some characters that displayed intrageneric variation, this approach was adopted to
minimise ambiguity in the application of generic names. In addition, two other plusiine taxa
were included: Palaeographa Kljutshko, which had been proposed only as a subgenus of
Syngrapha Hubner; and Autographa bractea ({Denis & Schiffermiller]), to represent the more
derived lineage of Autographa Hubner, a genus considered to be biphyletic by Eichlin &
Cunningham (1978) (but see Clade analysis).

Several plusiine generic level taxa were omitted. These are given in Appendix 2, together with
the reasons for their rejection. Unjustified emendations and subsequent misspellings are not
included. Several additional plusiines, other than the type species, were dissected for a variety of
reasons.

Antoculeora locuples (Oberthir, 1881): the BMNH series of Antoculeora ornatissima may include one or
more of this taxon, which were dissected in error. However, this did not affect the present study for the
putative A. /Jocuples specimen (1 ©’) did not differ from the putative A. ornatissima in any character
scored in this study.

Autoplusia olivacea (Skinner, 1917) (Fig. 81): dissected by chance from a mixed series with A. egena but
which proved useful in confirming several character states and transformation series.

‘Syngrapha’ gammoides (Blanchard, 1852) (Figs 82, 83): dissected for another study and used in the
present investigation because the male genitalia were less specialized than those of either A. egena or A.
olivacea and thus aided the clarification of several characters. ‘S.’ gammoides is almost certainly a species
of Autoplusia. This genus is currently being investigated by Fernando Navarro (Tucuman) and pending
the results of his study, I will refer to this moth as ‘Syngrapha’ gammoides.

Chrysodeixis argentifera (Guenée, 1852) (Fig. 40); C. illuminata (Robinson, 1968) (Fig. 41); C. kebea
(Bethune-Baker, 1906) (Fig. 42): the type species of Chrysodeixis, chalcites, has extremely modified
sternal hair pencils in the male (see Terminology) and it proved impossible to score characters 80—95 for
this species. Three further species of Chrysodeixis were dissected in order to determine the nature of the
modification in C. chalcites. C. kebea and C. illuminata were found to be almost as extreme as C. chalcites
but C. argentifera possessed a typical pair of sternal hair pencils. Characters 80—95 were thus scored from
this latter species.

Plusiopalpa shisa Strand, 1919: only a single male of the type species, P. dichora, was found in the BMNH
collection, and so a male of P. shisa was also dissected to check leg and abdominal characters. The

94 IAN J. KITCHING

identification of this specimen as P. shisa rested solely on its Taiwanese origin. A study of the genitalia
may demonstrate this species to be synonymous with P. adrasta (Felder).

In addition, 11 outgroup taxa were analysed (Appendix 1) and were chosen on the following
bases.

Stictoptera/ Lophoptera: to represent the two main lineages (Holloway, 1985) of the subfamily Stictopter-
inae, which is considered by many to be closely related to the Plusiinae. The BMNH series of the type
species of Lophoptera, L. squammigera, appears to be mixed (J. D. Holloway, pers. comm.) but this will
have no effect on the present study.

Paectes/Anuga: to represent the subfamily Euteliinae, considered by Holloway (1985) to be a more likely
candidate for the sister-group of the Stictopterinae. Paectes cristatrix was used inadvertently instead of
the type species, P. pygmaea Hubner.

Magusa: this genus superficially resembles Stictoptera and may be related to it.

Calophasia/Cucullia: to represent the subfamily Cuculliinae, which share with the Plusiinae the character
state of lashed eyes.

Diloba/Brachionycha: Diloba has been associated in the past with the Plusiinae (amongst many other
groups, see Kitching, 1984) and this relationship was to be tested. It has also been suggested that Diloba
is related to the cuculliine tribe Psaphidini and thus Brachionycha was included, as the only member of
this tribe sufficiently well represented in the BMNH collection to be utilized.

Chalcopasta/Oncocnemis: the former genus has a similar facies to members of the plusiine genera
Thysanoplusia and Diachrysia, and a relationship between the amphipyrine tribe Stiriini (of which
Chalcopasta is a member) and the Plusiinae has been very tentatively suggested. However, Hogue
(1963), in his revision of the Stiriini, strongly favoured a relationship between this tribe and the cuculliine
tribe Oncocnemidini, hence the inclusion of Oncocnemis as an outgroup in this study.

Several other genera were initially examined but were rejected because they appeared to
have no close affinity with the Plusiinae — Plusiophaes Prout, Scoliopteryx Germar, Xylena
Ochsenheimer, Lithophane Hubner, Nycteola Hiibner and Pseudoips Hibner. In addition, two
stictopterines, Odontodes Guenée and Nigramma Walker (now a synonym of Gyrtona Walker;
Holloway, 1985) were scored but rejected as redundant to the problem at hand.

All taxa were scored from specimens held in the BMNH collection except for the follow-
ing:

Loboplusia vanderweelei holotype C’, paratype O': Rijksmuseum van Natuurlijke Historie, Leiden,

Netherlands;

Autographa pokhara Dufay holotype 2 (= Loboplusia vanderweelei 2): Zoologische Staatssammlung,

Munich, West Germany;

Shensiplusia nigribursa holotype 9: Northwestern College of Agriculture, Wukung, China. This species
was not examined; however, as many characters as possible were scored from the original description

and accompanying diagrams of wing pattern and female genitalia (Chou & Lu, 1974).

Terminology

In this section, I shall define those morphological terms that may have ambiguous usage within the
Lepidoptera or that I have introduced in this study. They concern mainly internal structures, the male
sternal hair pencils and the male genitalia. Unambiguous or universally accepted terms are omitted.
Genitalic terminology generally follows Klots (1970); wing venation follows Neilsen (1985). Thoracic
segments are termed TI-III (Roman numerals) and abdominal segments A1—8 (occasionally —A10 with
reference to larvae). The abdominal tergites and sternites are termed T1—8 (note Arabic numerals) and
St1—8 respectively.

Wings. The scale tooth is a triangular tuft of scales projecting from the tornal angle of the forewing. In the
males, the frenulum consists of a single bristle formed by the fusion of several setae. In most of the species
examined in this study, the female frenulum consists of three setae. The innermost seta is termed 1, the
outermost 3 (Fig. 124). Reduction in number generally results from the loss of seta 1 first, and seta 2 next.
Whether reduction is by loss or by fusion is often difficult to assess. In those species with four frenular setae,
it is unknown whether the extra seta represents a ‘0’ or a ‘4’. Crossveins are identified by the main veins
between which they run, labelled in lower case. Thus m—mz is the cross vein between M, and M). The
terms lower, middle and upper discocellular veins are avoided.

SYSTEMATICS OF THE PLUSIINAE 95

Labial palps. The apical ‘pit organ’ of segment 3 is here termed ‘vom Rath’s organ’ (Rath, 1887), following
Kristensen (1984). The sensilla contained in the pit have been thought to be chemoreceptors, thermorecep-
tors or hygroreceptors (Chauvin & Faucheux, 1981), but have recently been shown to be sensitive to
carbon dioxide (Bogner et al. , 1986).

External thoracic structure. The tegulae (Figs 164-167) are interpreted as having two arms. The dorsal
arm is broad and usually bluntly pointed or rounded, and protects the forewing base dorsally. The ventral
arm is thin and covered in microspines apically. It curves out and under the wings, where its spinose apex
may aid in holding the wings in position at rest.

Legs. The basitarsal comb is a row (or rows) of specialized setae (Figs 178-181) on the inner subventral
edge of the hindbasitarsus of many male argyrogrammatines. The function of these combs is still unknown,
although their partial correlation with the dorsolateral hair tufts (see below) suggests that they may be used
for grooming. They do not seem to be absolutely essential, however, for although all species with combs
have tufts, not all species with tufts have combs. The genus with the most highly developed tufts of all,
Chrysodeixis, lacks combs.

In general, there are three discrete rows of setae along the ventral surface of the tarsi in plusiines. The
central row is termed the V (ventral) row, the lateral rows SV (subventral). Additional SD (subdorsal) and
D (dorsal) setae may also occur.

Internal prothoracic structure. The internal structure of the ventral part of the prothorax of a typical
plusiine is illustrated in Fig. 211. The ventral midline (discrimen) is inflected to form two large, curved
processes that form the dorsal arms of the prothoracic furca. Basally, these are produced posteriorly to
form a thin, apically bifurcate spinisternum (not shown). The dorsal furcal arms are produced subventrally
into broad, rounded projections termed the subventral (SV) furcal arms. To these structures are attached
several of the foreleg muscles. The apices of the dorsal arms are connected on each side to the dorsal edges
of the epimera by a cuticular membrane. This structure may possess a dorso-ventral ridge running from the
epimera to the SV furcal arms (Fig. 212).

Male abdomen A1-7. Terminology for the trifine hair pencil components follows that of Birch (1972a; b).
Dorso-lateral hair tufts occur in many male Argyrogrammatini. The tufts have greasy, matted, apparently
secretory patches of scales at their bases and are assumed to be scent organs, presumably involved in
courtship. Those of A7 are occasionally enclosed in cuticular folds of various types. All are easily removed
during normal preparatory procedure for slide mounting and then often go undetected.

Male A8 and sternal hair pencils. The modifications to A8, concerning the elaboration of the sternal hair
pencils of male plusiines, are very complex and still incompletely understood. The plesiomorphic condition
was presumably a simple quadrate T8 and St8. In many of the genera examined, including all of the
plusiines and several of the outgroups, T8 was cruciform, or bilobate, in which the anterior arms had been
reduced or lost. The shape of T8 is often species specific (Dufay, 1970a) but although much variation was
found in this study, no patterns to the different forms could be discerned and thus no characters were
derived from it.

St8 appears first to have evolved a central, posterior, desclerotized area giving rise to a U-shaped sclerite
in which the arms point posteriorly. Between these are developed a patch of secretory, setose scales. This
desclerotized area is then invaginated as two pockets (more rarely one), in which are located the sternal
hair pencils. These pockets are often fused, partially or totally, and the hairs themselves occasionally
differentiated into several types in terms of length, curvature, colour and basal insertion. The development
of the pockets bent the U-shaped St8 in the dorso-ventral plane, resulting in a thin strap supporting the
_ edge of the pockets ventrally, termed the anterior arm of St8, and two thin bars projecting posteriorly
above the opening of the pockets, the lateral arms of St8. The membrane between the lateral arms can be
variously scaled.

In the Argyrogrammatini, the lateral arms of St8 are further modified. Initially, they develop apical
secondary hair brushes. In some genera, the lateral arms also develop sub-basal, exterior lobes, which
represent the attachment points of the secondary hair brushes. The part of the lateral arm that is distal to
this lobe is referred to as the dorsal arm of St8, as it is a morphologically recognizable subdivision of the
lateral arm that lies dorsal to the hair pencils. In some argyrogrammatines, the dorsal arms bear flat plates
of greasy, closely appressed scales.

As alluded to above, many Chrysodeixis species have even more modified structures on A8. C.
argentifera is unmodified and is thus interpreted as being a relatively plesiomorphic species in terms of

96 IAN J. KITCHING

these structures. In other species, however, and especially C. chalcites and C. kebea, the sternal hair pencils
have been lost entirely and St8 modified into a compact sagittate sclerite (Fig. 250). The function of the
sternal hair pencils has been taken over by the secondary hair brushes, which have moved ventrally and
become retracted into a shallow, semi-circular pocket. T8 has also become highly modified into a U-shaped
sclerite with thin arms. It also has attenuated lateral and anterior arms that enclose and support the pocket
of the secondary hair brushes (Figs 251-253). In consequence, T8 in the ‘higher’ Chrysodeixis has become
remarkably convergent in form upon St8 in other argyrogrammatines, except that its components are
rotated through 180 degrees relative to those of St8.

The sternal hair pencils and associated structures of Trichoplusia ni have been described in detail by
Grant (1971) and their role in courtship was explained by Gothilf & Shorey (1976).

Tegumen and uncus. In most species examined in this study, there occur, latero-apically on the arms of the
tegumen, two small sclerites that project posteriorly and under the base of the uncus. These are termed the
‘latero-apical sclerites of the tegumen’ or LASTs. They may represent the remains of the gnathos since they
occur ‘ventro-laterad of the articulation of the uncus and socii if present’ (Tuxen, 1970).

Vinculum and pleurite. In this study, most of the species examined lacked a saccus in the male genitalia as
commonly understood. Instead, the arms of the vinculum converge to a point. In several genera, dorsal and
ventral flanges of these arms are produced and fused to form saccus-like structures but are referred to by
their appearance rather than as sacci.

The pleurite (termed the paratergal sclerite by Tikhominov, 1979) is a small sclerite, often curved and
flanged, that is found exteriorly and laterally at the point where the vinculum and tegumen arms meet. Ina
few genera, the pleurite was seen to be continuous basally with the outer surface of the vinculum and thus
appears to be an extension of the latter. In many plusiines, the base of the pleurite has become sharply
kinked and may have lost its connection with the vinculum. In a small number of taxa, the pleurite has
become fused with the dorsal edge of the tegumen or may have been lost altogether. In several genera, the
vinculum arms continue posterior to the origins of the pleurites and these extensions are termed the
ventral, apical, vinculum bars (VAVBs).

Valve. There is much confusion and ambiguity concerning the terminology of the valves and their
constituent structures. In this study, the following terminology is adopted. The valves are the paired,
presumably clasping organs of the male genitalia. The dorsal edge of the valve is the costa; the ventral,
often recurved edge, the sacculus. At the base of the sacculus is produced a usually tubular appendage
termed the clavus, while more distally, a similar structure is termed the harpe. The harpes are much more
variable in form than the clavi. The transtillae are sclerotized bars, produced from the bases of the costae,
that project inwards and support the anellar membrane dorsal to the aedeagus. The aedeagus is supported
ventrally by the juxta.

Aedeagus. The muscular tube arising from the base of the aedeagus, down which the spermatophore
passes, is termed the cuticular simplex (Callahan, 1960). If present, the enlarged section of the aedeagus
basal to the origin of the cuticular simplex, especially if bulbous, is termed the coecum. References in the
character descriptions of structures on the vesica as LHS (left hand side) and RHS (right hand side) are
from the moth’s point of view.

Female Al-7. All of the taxa examined in this study, in common with the majority of the rest of the
Noctuoidea, possess a counter-tympanal hood. This takes the form of a rounded, quadrate, sclerotized flap
produced from the pleural region of segment A2. It is concave on its inner surface and curves forward to
protect the metathoracic tympanal organ. The Euteliinae, Stictopterinae and Plusiinae were observed to
possess a second hood-like structure, antero-ventral of the counter-tympanal hood, which is often referred
to in the literature as a ‘secondary tympanal hood’. Close examination, however, shows that it is a flattened
pouch, the two walls of which are separate (Figs 230-231), not a single, solid structure, as is the
counter-tympanal hood. I will therefore refer to it as the ‘pleural pouch’.

The inner and outer surfaces of the pleural pouch are supported by two separate sclerotized bars of
markedly different origins. A latero-apical production of the outer margin of the anterior apodeme of St2
(Figs 230-231) projects antero-laterally and then recurves sharply postero-ventrally to support the inner
surface of the pouch. In contrast, the outer surface of the pouch is supported by a ventro-anterior extension
of the post-spiracular bar (Fig. 231).

In most euteliines, stictopterines and plusiines examined, the pleural pouch is a large structure,
projecting antero-ventrally below the counter-tympanal hood. However, in Polychrysia and its relatives

SYSTEMATICS OF THE PLUSIINAE 97

(see character 178), the pouch is small and has moved dorsally so as to be partially concealed by the
counter-tympanal hood (Fig. 232).

Female genitalia. The correct homologies of segment 8 of the female noctuid have yet to be fully
elucidated. For the purposes of this work, T8 will refer to the sclerotized semi-annulus encircling A8
dorsally and laterally. The often membranous section between the edges of T8 ventrally is termed the
lamella postvaginalis (LPV). The antrum is the initial differentiated section of the ductus bursae
immediately internal to the ostium bursae. It is frequently more heavily sclerotized and usually much
expanded relative to the rest of the ductus bursae. In general, the apex of the corpus bursae is the
posteriorly-directed section posterior to the entry point of the ductus bursae into the corpus bursae. The
fundus is the opposite, anterior end of the corpus bursae (but see also character 209).

Methods

For each species (except Plusiotricha, Eutheiaplusia, Stigmoplusia, Chrysanympha, Adeva and Loboplu-
sia), two specimens, one male and one female, were selected. First, general facies and pattern characters
were scored from the intact animals. Then, the right fore and hindwings were detached, bleached and
mounted in Euparal. From these, the wing structure characters were scored. Next, the head, prothorax,
patagia, tegulae and legs were removed, heated in 10% KOH solution, descaled and, except for the head
capsule and prothorax, mounted in Euparal. Most characters from these parts were scored before
mounting but several required confirmation by observation using a phase contrast compound microscope
at <400. Finally, the abdomens were removed, heated in 10% KOH solution and descaled, particular care
being taken with the males not to dislodge the hairs of the subdorsal scent patches or sternal hair pencils, if
present. As it was found that many genitalic characters would be obscured and/or distorted by the process
of mounting, they were scored from unmounted preparations and routinely checked by dissection of a
second example of each sex (except for Omorphina for which there was insufficient material). This was not
done for the other characters. However, in any instance where there was doubt or ambiguity as to the
correct coding of character states, additional specimens were examined. The numbers of each sex of each
species studied are given in Appendix 1.

The method varied slightly for the six genera mentioned above, for which only very limited material (2—6
specimens in total) was available. For five genera (not Loboplusia), the head and prothorax were not
removed. Only one tegula, patagium and set of legs were detached for examination. The wing venation and
structure characters were scored by wetting the wing in situ with ethyl acetate. The female frenulum was
examined by carefully brushing the scales away at the base of the wing underneath with a 00 sable-hair
brush. The genitalia of one male and one female only were dissected. For Loboplusia, only one set of legs
was removed from the paratype male for examination and the genitalic characters were scored from
pre-existing slide preparations. These measures resulted in numerous ‘missing data’ scores (*) in the data
matrix. In order to reduce their number and to speed up the computer analysis, any * character that was
found to be constant for all the other plusiine taxa was scored appropriately for the six rare genera. These
characters were mainly from the head, prothorax, patagia and tegulae. In addition, no females of any of the
three species of Plusiotricha (P. livida, P. carcassoni Dufay and P. fletcheri Dufay) are known and thus
characters 14, 176, 177 and 179-216 were scored as *. Character 178, scored from females in all other
genera, was scored from the male in Plusiotricha. Several characters also exhibited dependency upon the
possession of certain states for other characters. These non-comparable characters were also scored as *.
The data matrix is presented in Appendix 3.

Scanning electron micrographs of proboscis structures were obtained using samples heated in 10% KOH
solution, air-dried, affixed to stubs using an epoxy resin, coated with gold for 90s using a sputter coater
(Polaron E5000), and then observed using a Cambridge $180 scanning electron microscope, with an
accelerating voltage of 10 kV.

Characters

Where possible, the states of a character were arranged into an ordered transformation series.
The states of such a series are coded 0 (plesiomorphic), 1, 2. . . n (most apomorphic). For a
number of characters, there was uncertainty as to the most appropriate ordering of the states.
These characters are analysed as unordered (using the UNORDERED option of the PAUP
program; see also Analytical procedure), wherein no a priori assumptions of the order of state
transformations is made. Unordered characters are indicated below by an asterisk and the
coding of their states0,1,2. . . ncarries no connotation of polarity or transformation sequence.

98

IAN J. KITCHING

In several of the descriptions, the symbol + is used to signify ‘more or less’ or ‘approximately’.
Scale lines on the figures are all 1 mm unless otherwise stated.

Wing pattern

1 Raised scales in the forewing cell absent (0)/present (1).

2 Orbicular, suborbicular and reniform stigmata outlined in black, the former two partially or totally
fused (Figs 32, 33) (1)/stigmata not so formed (0).

3. Lower half of suborbicular stigma, with or without a basal, exterior spot, modified to form a Y-mark
(e.g. Figs 39, 43, 87) (1)/not (0).

4 *Forewing colour internal to the terminal line not pink (0)/varying from a few scattered pink scales to a
distinct pink line (1)/a pink line tending to be concentrated at the apex of vein M; (Fig. 53) (2)/a
conspicuous pink spot at the apex of vein M; (Fig. 54) (3)/a distinct white line (4).

5 Single black scales scattered over the wing surface (Figs 62, 71) (1)/not (0).

6 Scale tooth on the forewing tornus well developed (Figs 46, 86) (1)/weakly developed or absent (Figs 90,
92) (0).

7 Basal two-thirds of hindwing yellow (1)/grey or brown (0).

8 Base of hindwing upperside with broad scales (0)/hindwing base hyaline, scales setaceous (Figs 98, 99)
(1).

9 Forewing with large, discrete patches of green-gold metallic scales (Figs 47, 84, 85, 102) (1)/not (0).

10 Forewing with metallic sheen over parts, not arranged into discrete patches (1)/forewing without
metallic sheen (0).

11 Forewing with a conspicuous black spot in fringe at apex of M; (Figs 38, 39, 51) (1)/not (0).

12. Reniform stigma at least partially outlined in silver (Figs 58, 79, 92) (1)/not (0).

13 Forewing cell underside with hair-like scales (1)/without such scales (0).

Wing structure

14 *Structure of female frenulum:

(0): 1 < (2 = 3); 1 finer and just shorter than 2 and 3 (Fig. 124);

(1): 1 < (2 = 3); 1 is finer and half the length of 2 and 3 (Fig. 125);

(2): 1 < (2 = 3); 1 is very fine and almost lost (Fig. 126);

(SB) 2 = 3 (Fig: 120);

(4): 2 <3; 1 lost (Fig. 121);

(5): 3 only; I and 2 lost (Fig. 122);

(6): 1 =2 =3 = 4; 4 setae present (Fig. 123).
A single specimen of Pseudeva was observed with four setae on the right wing only, while a single
Omorphina was observed with five setae on each wing. All other examples of these genera examined
showed state 0.

15 Hindwing vein Mas strong as M; (Fig. 131) (0)/half the width of M; (Fig. 132) (1)/vestigial (no trachea
visible internally) (Fig. 133) (2)/lost completely (Fig. 134) (3).

16 Apex of forewing vein R: produced as a point beyond the apex of Rs and M, (Fig. 129) (1)/not (Fig. 130)
(0).

17 Forewing vein R> + 3 4 4 (upper edge of the areole) approximately parallel to R; (Fig. 135) (0)/markedly
converging on R, then curving away distally (Fig. 136) (1).

18 Hindwing vein M, arises near M; such that M, — M3; < M3; — CuA, (Fig. 134) (0)/M2 arises less than
halfway between M; and the bend in m; — m3, M2 — M; >> M3 — CuA, (Fig. 137) (1)/Mp arises half
way between M; and the bend in m; — m3, M2 — M; >> M3 — CuA, (Fig. 138) (2)/M> arises at the bend
in m; — m3, M, — M3 >>> M3— CuA; (Fig. 139) (3).

19 Hindwing vein M> apex produced as a blunt point, giving five concavities between R, and CuAp (Fig.
128) (0)/M> apex not produced, giving four concavities (Fig. 127) (1).

20 *Scale bases on wings randomly distributed (Fig. 140) (0)/arranged in irregular transverse rows between
the veins (Fig. 141) (1)/arranged in highly regular transverse rows, one scale wide (Fig. 142) (2). This
character is scored by observation of the scale base arrangement between forewing veins M3 and My.

21 Forewing vein CuP absent (0)/a distal short length of vein CuP present (Fig. 143) (1).

Head
22 Scales anterior to the bases of the antennae unmodified (0)/modified into lashes, not exceeding the eye

diameter (1)/lashes exceeding the eye diameter (2).

SYSTEMATICS OF THE PLUSIINAE 99

23 Eyes large and bulbous (Figs 95, 152) (1)/not (Figs 88, 148) (0).

24 Prominent rim round eye absent (Fig. 148) (0)/present (Fig. 151) (1).

25 Lower frons scaled only at the edge (1)/narrow, central band of frons, at most, unscaled (0).

26 Inlateral view, frons distinctly exceeding the eye (Figs 147, 148) (0)/frons barely exceeding the eye, if at
all (Figs 149, 150) (1).

27 *Pilifers elongate (Fig. 144) (0)/triangular (Fig. 145) (1)/very broadly triangular, not exceeding the
labrum (Fig. 146) (2).

28 Ventral to antennal sockets, frons slightly concave, so giving the appearance, in front view, of a
dorsally delimited frons (Fig. 147) (0)/no such concavity, frons appearing undelimited ventral to the
antennal sockets, sockets contiguous with it (Fig. 149) (1).

29 Postocciput flat or slightly concave (Fig. 149) (0)/distinctly convex and bulged (Fig. 147) (1).

30 In the male, frons approximately square, as long as wide (Fig. 151) (0)/distinctly narrower than long
(Fig. 152) (1).

Proboscis

31 *Proboscis distally armed with no styloconic sensilla (0)/few (Fig. 108) (1)/many (Fig. 109) (2).

32 *Proboscis long and thin (0)/shorter and stouter (1)/reduced and non-functional (2).

33 Styloconic sensilla with segment 1 cylindrical (Figs 110, 111) (0)/with five longitudinal ridges, apically
produced to give a stellate appearance when viewed end-on (Figs 112, 113) (1)/with the three distal
ridges produced as ‘wings’ (Figs 114, 115, 116) (2).

34 *Setae on outer surface of proboscis restricted to approximately the most basal half turn (0)/continuing
almost to the tip and the section with dense styloconic sensilla (1)/continuing to the apex (2).

35 *Strengthening bars of proboscis in the most apical part semi-circular (Fig. 108) (0)/broken up into
small, individual nodules (Fig. 117) (1)/as (1) but nodules only at the extreme apex (Figs. 109, 118) (2).
A nodular apex (state 1) is also found in many butterflies (Sellier, 1975) and probably represents the
plesiomorphic condition.

36 Basal microspines on proboscis irregularly and sparsely distributed (0)/arranged in regular, dense,
circumferential rows (Fig. 119) (1).

Labial palps

37. Ventral edge of segments I and 2 fringed with hairs (0)/fringed with broad scales giving a smooth edge
(1).

38 Lateral scales of segment 2 adpressed (0)/semi-erect (1).

39 Ventral fringe of segment 2 (and usually 3) formed into a neat ridge, giving a blade-like appearance
(1)/scales loose, edge not blade-like (0).

40 Segment 1 more than two-thirds length segment 2 (Fig. 155) (0)/segment 1 approximately equal in
length to segment 2 or slightly longer (Fig. 153) (1).

41 Segment 3 approximately as long as segment 2 (Fig. 156) (1)/segment 3 only two-thirds as long as
segment 2 or less (Fig. 155) (0).

42 Apex of segment 3 blunt (Fig. 155) (0)/acutely pointed (Fig. 154) (1).

43 Vom Rath’s organ conspicuously papillate basally (Fig. 158) (1)/not (Fig. 159) (0). All VROs are
papillate in that they have individual internal invaginations, presumably housing sensilla. However,
regular papillations, clearly visible under x 10 magnification, are termed conspicuous.

44 Segment 3 much longer than wide (Fig. 155) (0)/subspherical (Fig. 153) (1).

Antennae

45 Ventral hairs on male antenna short, less than width of the segments (Fig. 162) (0)/long and apically
very flexuous, much longer than segments are wide (Fig. 161) (1).

46 *Male antennae simple (Figs 43, 162) (0)/asymmetrically bipectinate (Figs 100, 161) (1)/symmetrically
bipectinate (Figs 92, 160, 163) (2).

47 *Pectinations of male antenna apically with a single small seta (Fig. 161) (0)/with a single large, stout seta
(Fig. 160) (1)/with two apical setae (Fig. 163) (2).

48 Longest pectinations on male antennae arising from base of the segments (Fig. 163) (0)/arising distally
on the segments (Figs 160, 161) (1).

External structure of thorax — scaling, patagia, tegulae
49 Large, bilobed, truncate crest present on TII (Fig. 39) (1)/absent (Fig. 104) (0).

100

50
51
52
53
54

55

56
37)

Legs

58
59

60

61

62

63

64

65

66

67

68

69

IAN J. KITCHING

Tegulae adpressed to TI (Fig. 72) (0)/tegulae held almost vertically away from TII (Fig. 62) (1).
Dorsal arm of tegula bluntly pointed or rounded (Fig. 166) (0)/truncate (Fig. 164) (1).

*Ventral arm of tegula bluntly pointed (Fig. 171) (0)/truncate (Fig. 172) (1)/spatulate (Fig. 173) (2).
Ventral arm of tegula with a colourless, microspinose extension (Fig. 169) (1)/not (Figs 168, 170) (0).
Microspines on ventral tegula arm restricted to apex and outer edge (Figs 168, 169) (0)/microspines
extending along inner edge towards main body (Fig. 170) (1).

Ventral arm of tegula produced on a lateral extension of the main body, outer edge of dorsal arm almost
evenly continuous with the dorsal surface of the ventral arm (Figs 165, 167) (1)/ventral arm not on a
lateral extension, outer edge of dorsal arm curves back dorsally to form a distinct ridge running roughly
parallel to the outer edge of the ventral arm (Figs 164, 166) (0).

Dorsal tegula arm with a broad marginal flange (Fig. 174) (1)/flange very narrow (less than I scale base
wide) or absent (Fig. 175) (0).

Attachment point of patagium drawn out (Fig. 176) (1)/not (Fig. 177) (0).

Foretibia with a dorsal white spot (Fig. 87) (1)/without such a spot (Fig. 84) (0).
Hind tibia and tarsi transversely banded grey, black or brown and white dorsally (Fig. 44) (0)/dorsally
unicolorous white, pale grey or pale orange (Figs 34, 87) (1).

*Basitarsal comb on male hindleg absent (Fig. 178) (0)/present as a single row of short setae
(approximately equal to the width of the basitarsus or to the length of the SV spines) (Fig. 179) (1)/a
single row of long, fine setae (approximately 2X the width of the basitarsus or the length of the SV
spines) (Fig. 180) (2)/as (2) but setae in multiple rows (Fig. 181) (3).

Inner row of SV spines on the fore-tarsi present in male on all segments along the entire length of each
(0)/inner row + suppressed over basitarsus (apical and subapical spines may remain) (Fig. 182)
(1)/inner row + suppressed on all segments, outer row suppressed on tarsi 2—5 (Fig. 183) (2)/all SV
rows suppressed, V row of spines on basitarsus short and stout (Fig. 184) (3).

Spines of hindbasitarsal SV and V rows equal along segment in both sexes (Fig. 185) (0)/spines on basal
half of segment smaller and finer (Fig. 186) (1)/spines suppressed on basal half of segment (Fig. 187)
(2). For the purposes of coding character 62, the male hindbasitarsal comb of certain plusiines has
been ignored.

*Fore- and midtibiae unspined (0)/with 1—2 apical spines on outer surface (occasionally I on foretibial
inner surface) (Fig. 195) (1)/fore- and midtibiae extensively spined on both surfaces (Fig. 196) (2).

* Hindtibiae unspined (0)/spined on outer surface distal to median spur insertions (Fig. 188) (1)/spined
along entire outer surface (Fig. 189) (2)/spined between the median and apical spur insertions
subventrally on inner surface (Fig. 190) (3).

*Tarsi with few small, SD and D setae (Fig. 194) (0)/with a small number of additional large spines, of
the same form as the SV and V spines (Fig. 193) (1)/with a few SD and D setae, finer than the SV and V
rows (Fig. 192) (2)/with numerous long, fine SD and D setae (Fig. 191) (3).

Hindleg (measured as femur base to apex of pretarsus) subequal to midleg or longer (0)/hindleg shorter
than midleg by approximately half the length of the mid-femur (1).

Claw with a ventral lobe (Figs 201, 202, 203) (1)/claw evenly curved ventrally (Fig. 204) (0).

Hindleg inner apical spurs less than half the length of the basitarsus (Figs 199, 200) (0)/approximately
three-quarters length of the basitarsus, occasionally subequal to it (Fig. 198) (1)/spurs as long as
basitarsus or longer (Fig. 197) (2).

Foretibia with a dorso-apical claw (Figs 205, 206, 207) (1)/claw absent (0).

Internal structure of prothorax

70

71

Dorsal arms of furca vertical, not apically produced (Fig. 209) (0)/dorsal arms produced and curved
posteriorly (Fig. 210) (1).

Dorsal edge of epimeron with a dorso-ventral ridge running from near the dorsal furcal arms to the
apex of the SV furcal arms (Figs 212, 213) (1)/without such a structure (Fig. 211) (0).

Male abdomen A1-—7

72
18)

Abdomen cylindrical or broadly tapered (0)/abdomen distinctly narrowly tapered, A7 half the width of
A2 (1).

* Internal flange on anterior edge of T2 curved and of uniform diameter (Figs 216, 219) (0)/with two SD
lobes (Fig. 220) (1)/lobes almost fused but with a central emargination (Fig. 221) (2)/flange semicircu-
lar, ventral edge straight (Figs 217, 218) (3). The flange in Cucullia is broad but uniformly so and with a

SYSTEMATICS OF THE PLUSIINAE 101

distinctly concave ventral edge thus distinguishing it from state 3 in which the ventral edge is straight.
The flange in Lophoptera is reduced but still has a straight ventral edge, as well as the characteristic
triangular dorsal inflection. The latter structure also serves to distinguish state 3 from extreme forms
of states 0 and 2.

74 Central membranous area of St2 flanked by grooves arising from anterior apophyses or by flat cuticle
(Fig. 222) (0)/flanked by two internal longitudinal flanges arising from the bases of the anterior
apophyses, and which are directed obliquely towards the midline (Fig. 223) (1).

75 *All components of trifine basal abdominal hair pencils absent (0)/some components present (1)/all
components present (2). Of the five main components of the trifine hair pencils (Birch, 1972a),
Magusa possesses only the lever and the pockets, and these are reduced and non-functional. Because
this condition could be the result of partial gain or loss, the character is treated as unordered.

76 St7 broad, + entire (Fig. 208) (0)/St7 with two postero-lateral sclerotizations, giving a ‘fish-tail’
appearance (Figs 214, 215) (1).

77 _ Dorsal-lateral hair tufts on A2—7 absent (0)/present on A5—7 (Figs 46, 57) (1)/additional ‘secretory’
areas present on A4, occasionally on A2—3 (2)/distinct additional tufts present on A4 (3). The secretory
areas on A2—4 are very friable and are easily lost during preparation. Thus no attempt was made to
subdivide state 2, especially as it would not be possible to reliably score Eutheiaplusia and
Plusiopalpa. Loboplusia has a unique distribution of what may be the scale bases of scent tufts on the
pleura of A2—4 but this is uncertain from the descaled preparation. The genus was thus coded as *.

78 *Scent hairs on A7 exposed (Fig. 226) (0)/enclosed in an invaginated groove (Fig. 227) (1)/enclosed in an
overfold of pleuron 7 (Fig. 228) (2)/enclosed in a membranous fold (Fig. 229) (3). These three states
are probably independent derivations.

79 *Dorso-lateral hair tufts on A5—6 with indistinct edges (0)/tufts on A5—6 with distinct, sharp antero-
medial edges (1)/as (1) but A5 only (2)/as (1) but A6 only (3).

Male St8 and the ventral hair pencils

80 *St8 with a thin anterior bar and two broad lateral bars extending posteriorly (Fig. 235) (0)/lateral bars of
St8 thin (Fig. 243) (1)/as (1) but anterior bar narrowed medially, often with latero-ventral, posteriorly-
directed sclerotized flaps (Fig. 254) (2)/lateral arm of StS produced at an acute angle to the anterior arm
(Figs 236-240) (3). This character probably comprises two independently derived forms of St8 and
hair pencil: the plusiine type, states 1 and 2; and the stictopterine type, state 3. The plusiine type is
probably derived directly from state 0 but this is much less certain for state 3. Subsequently, the two
types of hair pencil and associated St8s will be treated as mutually exclusive. Characters applicable to
one type will be coded in taxa showing the other type as *. The same coding will be used for those taxa
showing state 0 for 80. Several taxa exhibit partial reductions of the hair pencils and St8. Cucullia and
Abrostola have much reduced lateral arms but still can be interpreted as 80(3) and 80(1) respectively.
In the following hair pencil characters, 81—82 refer to the stictopterine type, while 83—95 refer to the
plusiine type.

81 *Hair pencils arising from a diamond-shaped pocket posterior to the anterior arm of St8 (Figs 238, 239)
(0)/from two shallow ventral pockets (Fig. 240) (1)/from two deep lateral pockets (Fig. 237) (2)/pencils
reduced to a ventral row of scales arising from a very thin sclerotized bar running just posterior to and
parallel with the anterior arm of St8 (Fig. 236) (3). These four forms of hair pencil appear to be
mutually exclusive.

82 Angle formed by anterior and lateral arms of St8 without an exterior lobe (Fig. 236) (0)/with a small
exterior lobe (Fig. 238) (1)/with a large, dorsally recurved bar (Fig. 239) (2).

83 *Hair pencils poorly developed, merely a tuft of sparse scales arising from a shallow depression (Fig.
241) (0)/hair pencil pockets slightly invaginated but still not discrete (Fig. 245) (1)/pockets well
developed, partially fused along their length (Fig. 243) (2)/pockets well developed, separate along their
entire length from St8 to their apices (Fig. 244) (3)/pockets well developed, totally fused into a single
structure but the two individual pencils still discrete (Fig. 265) (4)/single pocket present, elements fused
(Figs 242, 248) (5). It is uncertain whether the undeveloped hair pencils of Abrostola and of
Polychrysia and its relatives are plesiomorphically undeveloped or the result of subsequent reduc-
tion. All these genera were coded 0.

84 Hair pencil pockets with a sclerotized bar along the inner ventral edge (Figs 254, 258) (1)/absent (Figs
256, 263) (0).

85 Hair pencil insertions in one main apical field (Figs 255, 262) (0)/hair pencil insertions differentiated
with distinct, discrete subdorsal and lateral fields (Figs 258-260, 269-271) (1).

86 Hair pencil pockets with small subventral basal accessory pouches (Figs 260, 263) (1)/absent (Fig. 257)
(0).

102
87

88
89
90
91

92

93

94

95

IAN J. KITCHING

Hair pencils composed of scales of one length (Fig. 257) (0)/outer scales markedly shorter than inner
scales (Figs 263, 264, 267, 268) (1).

Hair pencils with an inner tuft that is much longer than the remaining inner scales of the pencil (Figs
260, 272, 273) (1)/all inner scales of approximately equal length (Fig. 266) (0).

Lateral arms of St8 bearing secondary hair brushes (Figs 242, 261, 267) (1)/not (0).

Lateral arms of St8 with an anterior flange where they curve dorsally over the hair pencils (Figs 260,
271) (1)/without such a flange (Fig. 264) (0).

Lateral arms of St8 undifferentiated, running to posterior margin of segment 8 as uniform linear bars
dorsal and lateral to the hair pencils (Fig. 243) (0)/lateral arms twisted dorsally over the hair pencils
(Fig. 257) (1)/as (1) but also with a deep basal concavity in outline (Fig. 262) (2).

“Lateral arms of St8 roughly linear (Fig. 244) (0)/with a broad triangular exterior sub-basal lobe (Fig.
256) (1)/lobe modified into a blunt projection (Fig. 258) (2)/lobe attenuate and recurved dorsally (Fig.
265) (3).

*Dorsal arms of St8 run parallel and do not meet apically (Fig. 245) (0)/dorsal arms converge to a point
(Fig. 254) (1)/dorsal arms run approximately parallel, fusing into a rounded apex (Fig. 263) (2)/dorsal
arms run transversely, approximately at right angles to the long axis of the body (Figs 258, 269) (3). The
dorsal arms are missing in Eutheiaplusia, the genus being coded * for those characters (93-94)
dependent upon their presence.

*Ventrally, dorsal arms of St8 unscaled or with unmodified scaling (0)/with broad, spatulate, erect scales

(Fig. 242) (1)/with broad, flat, ‘greasy-looking’ scales, appressed to the surface and adhering into sheets
(2)/with scales as in (2) but those apically on the dorsal arms formed into a semi-circular flat plate
projecting distally (Figs 258, 269) (3).
Membrane between lateral arms of St8 ventrally bare (0)/covered in long, fine, hair pencil-like scales
(1). These hairs are very easily brushed off, which may be what has happened in the genera
Antoculeora, Macdunnoughia and Allagrapha, in which the hairs are very sparse but nevertheless
recognizable.

Male genitalia — uncus, tegumen & anal tube

96

Dy
98
99

100
101

102
103
104

105

Uncus a simple, + cylindrical tube (Fig. 284) (0)/uncus flattened ventrally so it is D-shaped in
cross-section (Fig. 286) (1)/uncus reduced to a subspherical structure (Figs 276, 277) (2). This character
is treated as unordered. The three types are treated as independent for appropriate subsequent uncus
characters, as was done for hair pencils. It should be noted that there are numerous modifications to
state 0, some of which are coded below as separate characters.

Uncus basally curved (Fig. 282) (0)/straight (Fig. 283) (1). State 0 can show considerable variation
(Figs 281, 282, 284, 285).

Uncus of even width or gradually tapering distally (Fig. 284) (0)/uncus distinctly distally dilated
dorsally (Fig. 293) (1).

In lateral view, dorsal margin of tegumen + continuous with dorsal margin of uncus (Fig. 282) (0)/a
distinct notch present between tegumen and uncus (Fig. 285) (1)/apex of tegumen produced, apical
angle about 90 degrees (Fig. 287) (2). In state 1, the notch is formed by the downcurvature of the apex
of the tegumen and the increased basal curvature of the uncus. In state 2, the tegumen curvature is
accentuated to a sharp angle. Abrostola is superficially state 2 but the tegumen is apically rounded
(see Fig. 274). State 0 may be flat or have a uniform convex curve (Figs 281, 284). Several plusiines
(e.g. Trichoplusia, Dactyloplusia and Palaeographa) have a reduced state 1 that superficially looks
like state 0.

Base of uncus convex (Fig. 285) (0)/concave (Figs 274, 279) (1).

In lateral view, dorsal edge of tegumen, distal to pleurite, evenly curved or straight, with at most a short,
flat apex (Fig. 281) (0)/with a distinct angle, just over halfway along; when origin of uncus is horizontal,
the distal section of tegumen dips towards uncus base (Fig. 285) (1). Many other genera have flat distal
sections to the tegumen (e.g. Diachrysia, Chrysodeixis, Palaeographa) but these are all horizontal
when the uncus base is aligned as described above.

Uncus medially cylindrical (0)/medially slightly to greatly dorso-ventrally flattened and laterally
expanded (Figs 275, 278) (1).

In dorsal view, uncus with a distinct median suture visible, especially basally (Figs 275, 278) (1)/absent
(Figs 288, 290) (0).

In dorsal view, outer edges of tegumen diverge at about 90 degrees or less (Figs 288, 290, 296) (0)/outer
edges diverge at almost 180 degrees (Figs 295, 297) (1).

Dorsal edge of anal tube arising from base of uncus or extreme apex of tegumen (Fig. 282) (0)/anal tube

106

107

SYSTEMATICS OF THE PLUSIINAE 103

arising from bases of latero-apical sclerites of the tegumen (LASTs) to about halfway along the LASTs,
free of uncus base (Fig. 284) (1).

LASTs free apically of ventral edge of tegumen, projecting to articulate with the uncus base (Fig. 275)
(0)/LASTs fused to underside of tegumen, + fused dorsally around uncus base (Figs 290, 296) (1).
Also, in Paectes, the LASTs extend ventrally under and fuse under the anal tube as a sharply pointed
structure, a true gnathos. In their reduced form, the plesiomorphic state consists of a pair of lateral
bars extending from the ventral margin of the tegumen to the uncus base. In the plusias, the
membranous portion becomes sclerotized, either completely or leaving only a small, circular
membranous area basal to the uncus medially. In some weakly sclerotized genera (especially
Polychrysia and its relatives), there may be subsequent desclerotization but this is general and the
plesiomorphic condition is not displayed. Plusiopalpa dichora has extensive membranous areas
centrally but overall is like the other plusias. LASTs are absent in Anuga.

Dorsal surface of anal tube membranous (0)/with two subdorsal bars arising from the LASTs (1).
These bars are very weak in Abrostola. In Stictoptera, Lophoplusia, Lophoptera and Chrysanympha,
the connection with the LASTs is tenuous, while the bars in the latter two genera are fused into a
single dorsal bar. The bars in Mouralia bear subdorsal, wing-like projections.

Male genitalia — vinculum & pleurite

108
109

110

111

112

113

114

ils)
116

117

Vinculum arms with a small, distinct, ventral tooth (Fig. 303) (1)/without a tooth (Fig. 308) (0).

In lateral view, vinculum apically straight or with only a single slight curve (Fig. 299) (0)/vinculum with
two apical curves, dorsally then ventrally (Fig. 319) (1).

Arms of vinculum + straight in ventral view, vinculum V-shaped (Fig. 303) (0)/arms bent about
halfway along, vinculum Y-shaped (Fig. 301) (1). State 1 shows a distinct Y-shape, with the anterior
parts of the vinculum arms running + parallel. In Polychrysia and several other genera, the vinculum
is slightly Y-shaped but the distal parts still markedly converge.

Vinculum apically produced as a long, thin spine (Figs 300, 308) (1)/vinculum not so produced (0).
Both Pseudoplusia and Chrysodeixis have vincula produced to a sharp point. In Stigmoplusia, the
vinculum apex is rounded.

*Dorsally and subapically, dorsal flanges of the vinculum arms with even edges (Fig. 303) (0)/edges with
two small emarginations towards the exterior (Fig. 298) (1)/emarginations produced into two sharp
points (Figs 299, 300) (2)/emarginations produced into two blunt posteriorly-directed hooks (Figs 304,
305) (3).

Dorsal and ventral flanges of vinculum arms greatly expanded and flattened dorso-ventrally forming a
broad flat cone, unfused medially, sides of vinculum subparallel (1)/flanges of vinculum arms not so
enlarged (0). The apomorphic state for this character is a complex one that has to be qualified. The
subparallel sides are also found in those genera exhibiting state 1 of 110 but they do not have
expanded and flattened flanges. In Lophoptera (Fig. 309), the flanges are expanded and flattened,
but they are medially fused to form a complete cone and the edges are distinctly convergent.

Dorsal and ventral flanges of vinculum arms separate or, at most, ventral flanges fused (0)/dorsal and
ventral flanges fused across to give the appearance of a flattened cone (Figs 302, 309, 310) (1). In
Eosphoropteryx, the ventral flanges only are fused across. In Diloba, the dorsal flanges may be fused
but the general desclerotization makes this difficult to discern. The vinculum of Stictoptera is highly
derivative and although what is left of the flanges appear to be fused, a complete conical structure is
not found and the condition is not strictly comparable. Stictoptera was scored *. The fusion in Paectes
is drawn out but a conical form is still discernible.

Vinculum arms divergent to the base of the tegumen (Fig. 303) (0)/arms divergent and then run parallel
towards, or converge towards, the tegumen (Fig. 310) (1).

Pleurite dorsally simply concave, ventrally entire or with a desclerotized nick (Figs 317, 318)
(0)/pleurite dorsally strongly concave basally and then with a sharp convex bend caused by a distinct
sharp ventral basal notch (Fig. 312) (1). In Chalcopasta and Oncocnemis, there is a ventral basal
desclerotization but this is not manifested dorsally by a sharp bend. In Omorphina, there is a
dorso-basal bump but no ventral notch.

Pleurite without a lateral flange, lateral margin somewhat recurved towards the midline (Fig. 306)
(0)/pleurite with a lateral, often broad, flange, lateral margin flat (Figs 311, 319) (1). In Anagrapha and
Palaeographa, the lateral flange is somewhat reduced in width but the margin is nevertheless flat.
Pseudoplusia and Chrysodeixis have very thin pleurites with no lateral flange (or one that is very
narrow). The pleurite of Anuga is reduced to a small triangular lobe. In Paectes, a pleurite-like flange
is fused onto the antero-dorsal margin of the tegumen (see character 123), while in Diloba, a thin
sclerite, presumably the pleurite, occupies a similar fused position. In the stictopterines, the pleurite

104

118

119

120

IAN J. KITCHING

is either reduced to a small lobe (Stictoptera) or lost (Lophoptera), or fused with the tegumen and the
vinculum connection lost.

Apex of pleurite recurved dorsally over the tegumen to articulate on the inner edge in a pocket (Fig. 314)
(0)/pleurite apically lying dorsally on the tegumen or slightly recurved over the inner edge, but not
articulating in a pocket (Fig. 313) (1)/tegumen rotated almost vertically, pleurite apex only in contact
with the tegumen, otherwise lying parallel with the ventral apical vinculum bar (VAVB) (Fig. 316) (2).
In state 2, the pleurite and the VAVB appear as linear continuations of the vinculum apex.

Pleurite and VAVB fused basally (Fig. 312) (0)/fused to about half the length of the pleurite (Figs 306,
307) (1).

Pleurite broad, curved in lateral view and separate from the antero-dorsal margin of the tegumen (Fig.
312) (0)/pleurite very narrow, very closely appressed to (but not fused with) the antero-dorsal margin of
the tegumen (Fig. 289) (1).

121 *VAVB a long linear bar projecting from the apex of the vinculum to articulate ventro-apically on the

122

123

tegumen, apex of vinculum and base of tegumen distinctly separate (Fig. 312) (0)/VAVB very short or
absent, tegumen and vinculum closely approximated (Fig. 311) ()/VAVB absent or reduced to a blunt
lobe, tegumen and vinculum separated (Figs 318, 321) (2)/VAVB absent, tegumen produced basally
below pleurite to approach vinculum apex (Figs 315, 317) (3). In state 1, if the VAVB is present, then
it is bent, thus allowing the vinculum and tegumen to come into close proximity; in state 0, the VAVB
is straight.

VAVB produced ventro-posteriorly into a large, articulated socius (Figs 291, 292) (1)/not (0). In
Paectes, the socii are hairy, unsclerotized, blunt lobes; in Anuga, they are long, sclerotized, curved
and apically hooked, and are remarkably convergent in appearance upon the typical plusia uncus.
Vinculum and tegumen separate or linked by the VAVB (0)/vinculum and tegumen broadly joined,
pleurite totally fused along the antero-dorsal edge of the tegumen (Fig. 320) (1). This character could
also be represented as yet another alternative state of character 121. However, given that the
euteliines and stictopterines have been tentatively coded for character 121 (as opposed to * for
Diloba), characters 121 and 123 have been retained as separate.

Juxta & anellus

124

125

126

127

128

129

130

Anellar membrane present between ventral margin of aedeagus and apex of juxta (Fig. 323) (0)/apex of
Juxta + fused to apex of aedeagus ventrally (Fig. 322) (1).

Apex of juxta uniformly thin (0)/apex with a transverse apical thickening and a median longitudinal bar
(Figs 324, 325) (1). The longitudinal bar in Abrostola is weak, while in Mouralia, it is apparently
double.

Juxta free from valve latero-basally (0)/juxta fused to inner edge of sacculus near the base of the clavus
by a short thin bar (Figs 324, 325) (1).

Apex of juxta entire or if decurved, no notch or apical process present (Figs 325, 329) (0)/apex folded
ventrally to form a + parallel-sided notch that may be partly sclerotized dorsally to form an apical
process (Figs 326, 327) (1)/process well sclerotized distally so as to appear subapical on the juxta (Figs
335, 336, 340) (2)/process further modified into a sharp upcurving spine (Figs 337, 338, 339) (3).
Thysanoplusia, Zonoplusia and Ctenoplusia have juxtas in which the apex comes to a point without
curving ventrally. Similarly, in Trichoplusia, Acanthoplusia, Agrapha and Anadevidia, the juxta is
apically deeply concave but not decurved. Also, in these six genera, the anellus arises apically from
the juxta. In those genera showing states 1-3, the anellus arises dorsally and subapically from the
juxta, due to the dorsal sclerotization of the apical process.

Apex of juxta smooth (0)/apex of juxta scobinate (Figs 328, 329, 330, 334) (1)/scobinations distally
modified into sclerotized thorns (Fig. 341) (2). The scobinations in Chrysodeixis are very weak, while
in Pseudoplusia, they are restricted to the lateral, distally-produced arms of the juxta. Thus, the
states in these two genera may not be homologous with that in Dactyloplusia and Thysanoplusia. The
apices in Plusiopalpa and Cucullia are also heavily scobinate and even spinose (Plusiopalpa shisa),
but in these genera the juxta merges gradually into the anellus and so it is difficult to say on which
structure the ornamentation is located. Nevertheless, they were both coded 1.

Juxta a + quadrate sclerite, and usually well sclerotized (0)/juxta crescent- or boomerang-shaped,
poorly sclerotized (Fig. 333) (1). The juxta of Polychrysia and its relatives is poorly sclerotized but
distinctly quadrate. Stictoptera has a highly autapomorphic juxta and is coded *. The juxtas of
Brachionycha and Diloba are somewhat crescentic but are well-sclerotized.

Juxta basally separate from saccular flanges (Fig. 342) (0)/juxta basally fused to basal saccular flanges
(Fig. 343) (1). Those genera lacking saccular flanges (see character state 131:0) are coded *.

SYSTEMATICS OF THE PLUSIINAE 105

Valve

Lea
132

133
134

135

136
137

138
139

140
141

142

Base of sacculus evenly rounded (Figs 372, 373, 374) (0)/with a basal flange (Figs 351-355) (1).
Saccular flanges fused across and produced medio-anteriorly into a rounded lobe between the valve
bases (Fig. 343) (1)/flanges not so modified (0). In most genera with saccular flanges, they are either
separate from each other or fused across to form an irregular flat area of sclerotization. In two genera
(Panchrysia and Chrysanympha), however, the flanges are formed medially into a rounded,
somewhat dorsally concave lobe, giving, with the valve bases, a regular anterior outline like a
rounded M or a two-cycle sine wave.

Ventral edge of valve basally with a single row of enlarged flattened blade-like setae (Figs 351, 355)
(1)/without such setae (0).

Sacculus, distal to clavus, produced into an irregular spatulate lobe (Figs 347, 348) (1)/not (0). This
structure has been widely interpreted in the past as the clavus. However, its position (more distal on
the sacculus margin than the position occupied by the clavus in other noctuids) and the presence of an
obvious clavus in the ‘correct’ location in Mouralia argue that the spatulate lobe is not the clavus but a
novel development apomorphic for the Abrostolini.

Apex of valve with a distinct uniseriate corona (Fig. 349) (1)/absent (0). The corona is defined here as a
distinct marginal series of strong basally-directed spines. In Calophasia and Chalcopasta, the corona
is weak and somewhat submarginal.

Apex of valve not differentially scaled (0)/with multiple rows of easily detached, basally-directed, setose
scales; when removed, scale bases forming a distinct field (Figs 351-355) (1).

Marginal setae on ventral edge of valve on normal pinacula (Figs 347—350) (0)/some setae arising from
marginal prominences (Figs 368, 369, 370) (1). The marginal prominences are much larger than
normal pinacula and frequently bear two or more setae.

Valve constricted at about two-thirds of the distance base to apex, apical section spatulate (Figs 354,
355) (1)/valve not constricted (Figs 379, 382) (0).

Dorsal edge of saccular overfold with a lobe that overlies and obscures the clavus base when the valves
are opened flat (Figs 360, 368, 369, 370) (1)/clavus base, if present, not obscured by sacculus when
valves opened flat (Figs 348, 361) (0).

Ventral edge of valve basally evenly curved (0)/with a broad, blunt point about one-third from base
(Figs 351, 352) (1).

Sacculus with a broad, triangular spine basal to the origin of the harpe (if present), often further
modified (Figs 370, 376, 378) (1)/absent (0). This saccular spine is modified into a quadrate lobe in
Erythroplusia and into a long, serrate, lance-like projection in Antoculeora.

Apex of sacculus continuous with ventral margin of valve (0)/apex produced as a sheet, costa extended
ventrally towards it, but not fused with it (Figs 368, 369, 370) (1)/saccular sheet and costal overfold
fused, often modified into a curved hook (Figs 361, 372, 373, 374, 376) (2). This character is highly
problematical given the high degree of structural diversity displayed by the genera that possess it.
Allagrapha and Macdunnoughia exhibit the simplest form in which the apex of the sacculus is
extended beyond the ventral margin of the valve as a broad, blade-like sheet. This is modified in
Puriplusia into an inwardly-curving hook. In Autoplusia, a condition similar to that seen in Puriplusia
is found, in that the apical saccular process is modified to form an inwardly-curving hook. However,
in Autoplusia, there is a distinct fusion with the costal margin in the formation of the hook that does
not appear to occur in Puriplusia. Sclerogenia is similar to Autoplusia but the hook is larger, apically
blunt, and is curved ventrally rather than inward, while the ventral edge of the valve is greatly
enlarged, giving the hook an apparently subdorsal origin on the valve. The valve of Antoculeora is
highly autapomorphic and very difficult to homologize with the valves of the other genera. I have
chosen to interpret the ventral valve digitations, and in particular the subapical one, as being
homologous to the saccular hook of Autoplusia and Sclerogenia. However, given that the inner
margins of the costa and sacculus in Antoculeora are completely fused along almost their entire
length, a more comprehensive study is required to confirm this and the previous interpretations of
this character.

143 * Median edge of sacculus basally unmodified (0)/produced into a broad, rounded, setose lobe (Fig. 356)

144

(1)/lobe elongated into a distinct clavus, dorsal and ventral margins of which are basally + level (Fig.
357) (2)/dorsal margin much cut back (or ventral section basally elongate) so that clavus appears to be
located on a sclerotized peninsula (Fig. 358) (3).

Clavus a distinct tubular structure, >3 xX as long as basally wide (Fig. 360) (0)/clavus a short tubular
structure, aboutas long as basally wide (Fig. 380) (1)/clavus reducedtonomorethanaslightsetosebumpon
the end of the peninsula (Fig. 382) (2). Thischaracter and 145 are dependent uponstates of character 143.
Character 1441s coded * for taxashowingstatesOand | forcharacter 143, while character 145iscoded* in

106

145

146

147

148

149

150

151
152

153

154

155

IAN J. KITCHING

taxashowing states 0, 1 and2 for character 143. Character 144 isincluded because itis one of the ‘classic’
plusiine characters and is the one used by McDunnough (1944) to divide the plusias into three groups of
genera. Cornutiplusia and Polychrysia both have fairly short clavi but still long enough to be coded 0.
Similarly, Euchalcia and Lamprotes are somewhat intermediate between states 1 and 2 but given that
some tubular structure can be detected, they were coded 1.

Clavus borne ona short peninsula (Fig. 358) (0)/clavus borne ona highly elongate, somewhat flattened
peninsula that is about as long as the tubular clavus (Fig. 359) (1). Genera possessing short or very
reduced clavi (character 144: 1 & 2) are not strictly comparable and were coded 0, since in these
genera the peninsula is not elongated, but is longer than the clavus by virtue of the latter’s reduction.

Valve overall blade-shaped, ventral edge + straight for about two-thirds of its length, then curved up to
the apex of the costa to form a blunt or sharp point (Fig. 381) (1)/apex of valve + evenly rounded (0).
The shape of the valve coded by state 1 is readily recognized but difficult to describe and delimit
briefly. The general impression is of a knife blade; the valve gradually widens from the base, the costa
is slightly concave and the ventral margin possesses well-developed, sclerotized setae. In Cucullia,
the dorsal and ventral edges of the valve are + parallel and the genus was coded 0. The valves of
Diachrysia chryson are less blade-like than most genera coded 1, but other species in this genus,
notably D. nadeja, are as extreme as the autographas. Anagrapha is interpreted as state 1, due
allowance having been made for the highly modified, excavated apex in this taxon.

Valve + oval in outline, roughly symmetrical along the midline (Fig. 350) (1)/valve not oval,
asymmetrical along the midline (0). Euchalcia variabilis has a somewhat quadrate valve but is still
coded 1. Most other Euchalcia species have simple, oval valves (Dufay, 1968). Note that states 1 for
characters 146 and 147 are mutually exclusive. Taxa showing state 1 for one character are coded * for
the other.

Harpe present (1)/absent (0). If the saccular processes Autoplusia, Sclerogenia and Antoculeora are
reinterpreted as harpes, then states 1 and 2 of character 142 are not part of the same transformation
series.

Harpe a blunt, tubular process (0)/harpe modified into a sharply pointed, basally curved hook (Figs
354, 367, 379, 380, 386) (1). A number of genera have non-tubular harpes (e.g. Lophoptera,
Calophasia, Macdunnoughia) that are also not of the form of a curved hook (see also character 150).
Because they do not show the apomorphic condition of character 149, these genera will be coded 0,
even though they do not possess the plesiomorphic condition for this character either. The same
rationale applies to the following character also. The hooked harpes of Syngrapha, Caloplusia and
Palaeographa are almost certainly convergent upon those of Trichoplusia, Dactyloplusia and
Agrapha, but a wider survey of species needs to be performed before the two (or more) forms here
coded as state 1 can be adequately distinguished.

Harpe a blunt, tubular process (0)/harpe a strongly dorsally and inwardly curving hook (Fig. 369)
(1)/base of harpe also with an upcurving shorter hook forming a pincer-like structure (Fig. 370) (2).
This hooked modification of the harpe curves towards the midline when the valves are closed,
whereas that in character 149 curves in the plane of the valve towards the valve base.

Valve apically entire (0)/valve apically deeply cleft, costa and sacculus separate (Fig. 387) (1).

Valve much longer than wide (0)/valve + as long as wide, overall outline somewhat trapezoid (Fig. 375)
(1).

When closed, valves lie in a lateral position, easily opened flat (Fig. 389) (0)/when closed valves lie
ventrally or subventrally, difficult to open flat (Fig. 388) (1).

Transtillae fused dorsally above aedeagus and produced posteriorly as a broad, blunt, triangular point
(Fig. 346) (1)/transtillae separate or, if fused, not produced as a broad, blunt, triangular point (Fig.
344) (0). The euteliines lack transtillae and are coded * for this and the next character.

Transtillae a single bar, though often curved, angled or broadly triangular (Fig. 344) (0)/transtillae
laterally bifurcate into two thin bars diverging approximately at right angles; one passes dorsally over
the aedeagus and fuses with its opposite number, the other runs lateral to the aedeagus (Fig. 345) (1).
Omorphina has thin, fused bars passing over the aedeagus but no lateral bars. Allagrapha has broad,
ill-defined lateral bars that may be homologous to the thin, heavily sclerotized, well-defined bars in
Sclerogenia and Erythroplusia, but more species need to be investigated to confirm or refute this
possibility.

Aedeagus

156

Aedeagus large and thick (0)/aedeagus very thin and small, relative to the rest of the genitalia (1). This is
a ‘Gestalt’ character, very difficult to define objectively but easy to recognize in the apomorphic
state.

157

158

159
160

161
162

163

164
165
166

167

168
169

170

171

172

173

SYSTEMATICS OF THE PLUSIINAE 107

*Cuticular simplex arising dorso-basally on aedeagus, coecum + absent (Fig. 390) (0)/cuticular simplex
arising distinctly sub-basally, coecum distinct and of same diameter as aedeagus (Fig. 391) (1)/coecum
distinct and bulbous (Fig. 392) (2). Insome plusiines, the bulbous coecum is only just discernible as a
slight increase in the diameter over the apical diameter (e.g. Antoculeora). In those genera exhibiting
state 0 (e.g. Oncocnemis), the aedeagus in ventral view is of constant diameter (Fig. 393) and in
lateral view, the ventral basal curvature is often more gradual than the dorsal basal curvature (Fig.
390) (e.g. Cucullia).

Aedeagus generally sclerotized (Fig. 393) (0)/sclerotization apically restricted to a ventral band that
gradually broadens onto the coecum (Fig. 394) (1)/sclerotization restricted to a thin ventral band along
the entire length of the aedeagus distally, broadening somewhat abruptly onto the coecum (Fig. 395)
(2). Also, the ventral band of state 2 is extremely well defined laterally, the edges being straight and
parallel. The rest of the aedeagus is entirely membranous. In state 1, the ventral band merges
somewhat into the membranous area laterally.

Vesica + covered for entire length in small, basally-directed spines (1)/vesica not so ornamented (1).
When inverted, vesica very strongly and distinctly serpentine (Fig. 406) (1)/vesica not folded thus when
inverted (0).

Base of ductus ejaculatoris with a small, oval, sclerotized plate (1)/without (0).

Vesica with an isolated, subapical cornutus (SAC) (1)/without (0). Paectes has two cornuti, one of
which could be interpreted as subapical but, as the form of the vesica is short and inflated, rather than
long and tubular, as in the plusiines, Paectes was coded 0. Caloplusia has a spinose, sclerotized plate
in the approximate position of a SAC but was coded 0. Autoplusia bears a large, lateral cornutus
about halfway along the vesica that could be interpreted as a SAC. This would, however, require that
either the SAC has moved from a subapical to a median position on the vesica or that a large portion
of the ductus ejaculatoris has become inflated and gained a large field of numerous small cornuti. It is
more parsimonious to interpret Aufoplusia as having no SAC. Similarly, Palaeographa has a
recurved hook about halfway along the vesica that also is not homologous to a SAC. A SAC was
reported and illustrated in Eutheiaplusia by Dufay (1970a) but was absent from the specimen
examined in this study. The genus was coded 0. SACs were also occasionally absent in specimens of
Zonoplusia and Polychrysia. In Rachiplusia, the SAC is double.

SAC apically swollen (Figs. 408, 409) (1)/apically tapered (Fig. 403) (0). In Adeva, the SAC is short
and therefore somewhat stubby. It is nevertheless not apically swollen. Taxa coded 0 for character
162 were coded * for characters 163-166.

SAC arising from a circular, concave plate that has a somewhat irregular margin (Figs. 362-365)
(1)/SAC arising from vesica without a differentiated, plate-like base (0).

SAC arising apically from a lateral diverticulum (1)/SAC arising directly from the main vesica tube (0).
The SAC of Plusia arises at the base of a diverticulum and was coded 0.

SAC tubular (Fig. 403) (0)/SAC dorsally grooved (Fig. 404) (1)/edges of groove produced dorsally as
‘wings’ (Fig. 405) (2).

Vesica with a membranous sac invaginated from the apex of the basal RHS vesica sclerotization and
projecting anteriorly within the vesica (Figs 397, 401) (1)/vesica without such a sac (0). This sac is the
‘bulbous lock’ of Callahan (1960). In Trichoplusia, Thysanoplusia, Argyrogramma and Stigmoplu-
sia, the sac is small and + fused to the vesica sclerotization internally. In Plusiotricha, the sac is very
long, reaching internally almost to the apex of the coecum.

Base of vesica membranous and smooth (0)/internally granular, often forming a distinct annulus (1).
Basal vesica sclerotization apically confluent with the membranous part of the vesica (Fig. 397)
(0)/basal sclerotization produced ventrally as a blunt, apical point (Figs 398, 399) (1)/apical point
elaborated into a large, blunt, recurved hook (Fig. 407) (2). The recurved hook in Abrostola is
subventral on the LHS, rather than ventral but is interpreted as a further development of the ventral
blunt point found in Mouralia.

Vesica with a large, RHS, membranous diverticulum distal to the basal sclerotization bearing a field of
cornuti on its outer-distal surface (1)/without such a diverticulum (0). In Mouralia, the field of cornuti
is reduced to a small, sclerotized plate with several short cornuti.

Vesica sub-basally with a dorsal, isolated patch of broad, curved, short, pointed cornuti (Fig. 396)
(1)/without (0). In Lamprotes, the cornuti are connected to a LHS, lateral, sclerotized bar. Similar
features are found in other genera (e.g. Antoculeora, Sclerogenia), but the location and apparent
origin do not appear to be the same as for the cornuti here coded by state 1.

Vesica with two short, straight, pointed cornuti dorsally distal to the apex of the basal sclerotization
(Figs 397, 401) (1)/without (0).

Basal sclerotization of vesica on RHS uniform (0)/modified and reduced to a lateral, longitudinal band,

108

174

175

IAN J. KITCHING

differentially sclerotized dorsally and ventrally (Fig. 400) (1). Plesiomorphically, the base of the
aedeagus appears to have been + uniformly and heavily sclerotized, forming a basal, permanently
everted annulus. In many plusiines, this basal sclerotization has been reduced to two lateral bars, one
(commonly the LHS) or both of which are also frequently absent. In the Argyrogrammatini, the RHS
bar represents the only basal sclerotization of the vesica and is inverted when the vesica is inverted
and retracted. Further, this bar has heavier dorsal and ventral sclerotization, which gives it a grooved
appearance, and in extreme instances (e.g. Plusiopalpa), the RHS bar appears to be longitudinally
split. Some other species also have a RHS bar as the only basal sclerotization (e.g. Rachiplusia) but in
these taxa, the bar is uniformly sclerotized. In Zonoplusia, the sclerotization is reduced to two, short,
rudimentary bars, while the dorsal part is incomplete basally in Eutheiaplusia. The bar is very thin in
Stigmoplusia but the differential sclerotization can be discerned in the somewhat spatulate apex.
Basal RHS lateral sclerotization of vesica produced and turned dorsally to form a smooth, convex,
rounded plate (Figs 397, 401) (1)/not (0).

Sub-basal RHS auricular cornutus present (Figs 383, 384, 385) (1)/absent (0).

Female abdomen A1—7

176

ia

178

Pleural membrane of A7 thin, as on A2—6 (0)/pleural membrane of A7 thickened, white, apparently
glandular and often produced into a shallow pocket (1). The A7 pleural membrane in Puriplusia is
only slightly thickened and apparently glandular. The thickening in Allagrapha is restricted to a
shallow pouch on the posterior margin of A7. Autoplusia possesses a shallow pouch also, which is
produced into a deep, conical invagination in ‘Syngrapha’ gammoides. Loboplusia has a pleuron 7
that appears to be glandular but is not thickened. It is, however, difficult to be certain, so the genus
was coded 0. In one of the two specimens of Brachionycha, pleuron 7 is thickened and white but is not
apparently glandular. Nevertheless, the genus was coded 1.

Lateral edges of T7 convex or + straight (Fig. 224) (0)/concave (Fig. 225) (1). This character is almost
certainly correlated with the last. The lateral margin of T7 in Allagrapha is only slightly concave
(probably the result of the thickened area of pleuron 7 moving to the posterior margin of the
segment), while in Puriplusia, the margin is straight (probably due to the pleuron in this genus being
very slightly modified). The lateral margins of T7 in Loboplusia appear concave and the genus was
scored 1, although the condition is somewhat uncertain. If the lateral margins are concave, and
characters 176 and 177 are correlated, then it would be predicted that the examination of fresh
specimens of Loboplusia would reveal, at least, a glandular, slightly thickened pleuron 7.
Post-spiracular bar and anterior apodeme of segment St2 unmodified, pleural pouch absent (Fig. 234)
(0)/post-spiracular bar and anterior apodeme of St2 modified to support a large pleural pouch (Figs
230, 231, 233) (1)/as (1) but pleural pouch small and partially concealed by the tympanal hood (Fig.
232) (2). This character was frequently unobservable in many specimens due to loss of the relevant
parts on removal of the abdomen or during subsequent preparation. Adeva, Polychrysia, Eosphor-
opteryx, Chrysanympha, Lophoplusia and Paectes were scored, by necessity, from male specimens.
Chalcopasta lacks a post-spiracular bar.

Anal papillae & A8—A9 + 10 intersegmental membrane

179

180
181

182

183

Anal papillae + square or tapered apically, inner and outer surfaces parallel in dorso-ventral plane
(Fig. 414) (0)/anal papillae flared apically, inner surface partially or fully facing posteriorly (Figs 417,
418) (1).

Inner surface of anal papillae membranous and devoid of setae (except for the extreme apex) (Fig. 414)
(0)/inner surface densely setose (Figs 417, 418) (1).

Apex of anal papillae entire (Figs 414, 417) (0)/apex with fine, irregular, setose prominences, giving a
frayed appearance (Fig. 418) (1).

Anal papillae setae long, fine and undifferentiated (Fig. 420) (0)/anal papillae with a dorso-ventral line
of strong, pale-socketed setae, one-third to half way from the base, remaining setae mostly small and
fine (Figs 414, 415, 416) (1). Oncocnemis has a central area with very stout, bluntly pointed setae
sub-basally and a subapical row of similar but recurved setae. However, this setal type is autapomor-
phic for the genus and, because the basal and apical setae are of the long, fine type found in the other
outgroup genera, Oncocnemis was coded 0.

Anal papillae + square, longer than wide or highly elongate (Figs 414, 415, 416) (0)/anal papillae short,
distinctly wider than long (Fig. 419) (1). The plesiomorphic condition encompasses all anal papilla
forms from the square, somewhat apically lobed type of most plusiines to the highly elongate,
pointed type of Brachionycha and Chalcopasta. The form found in the euteliines and stictopterines

184

185

186

SYSTEMATICS OF THE PLUSIINAE 109

has been coded separately above (character 179) and the genera of these subfamilies are coded * for
character 183.

Membrane between A& and anal papillae dorsally extended as a pouch (Fig. 412) (1)/membrane not so
modified (Fig. 413) (0). The presence of the pouch has the effect, when everted and extended, of
turning the anal papillae ventrally. In addition, the pouch may be strongly evaginated but this state
proved too difficult to score consistently and was not employed.

Base of posterior apophysis dorsally membranous or with a sclerotized flange indistinctly differentiated
from the base of the anal papilla (Figs 416, 420) (0)/dorsal flange of posterior apophysis sub-basal,
distinctly separated from the base of the anal papilla and often elongate and leaf-like (Figs 414, 415) (1).
Many plusiines possess a basal flange on the posterior apophysis but only in the Argyrogrammatini is
it distinct from the base of the anal papilla. In other genera (e.g. Allagrapha, Puriplusia, Lophoplu-
sia, Anagrapha), the flange is an irregular sclerotized patch in the angle between the anal papilla and
the apophysis or on the apophysis but connected to the base of the anal papilla by a semi-sclerotized
zone of granular membrane.

Membrane between lamina post-vaginalis (LPV) and anal papillae unmodified, extension of anal
papillae ventrally easy (0)/membrane between LPV and anal papillae short, thickened and invaginated,
preventing protraction of anal papillae ventrally (1).

A8, ostium bursae & antrum

187

188

189

190
191

192

193

T8 uniformly sclerotized laterally (Fig. 424) (0)/T8 with a subdorsal, longitudinal, desclerotized band,
the ventral edge of which is continuous with the dorsal edge of the anterior apophysis (Figs 422, 423)
(1). Chalcopasta also has a desclerotized band but this has its apex along the midline of the anterior
apophysis, so that the ventral edge of the desclerotization is ventral to, not continuous with, the
dorsal edge of the anterior apophysis. In Lophoptera and the euteliines, the anterior apophysis is
absent (see 190). All of these genera were coded 0.

Anterior edge of T8 ventral to anterior apophysis flat (Fig. 424) (0)/invaginated as a dorso-ventral
groove, with a ventral pocket (Figs 421, 422, 423) (1). The ventral pocket is very small in Mouralia but
is enlarged, inflated and somewhat recurved in Abrostola.

T8 irregularly cut back antero-subventrally below the origin of anterior apophysis, leaving the ventral
base of anterior apophysis ‘free’ (Fig. 425) (1)/T8 not cut back antero-subventrally, base of anterior
apophysis ventrally continuous with anterior edge of T8 (0). State 1 for this character is variable in its
expression. Antoculeora was coded 0 although its genitalia are so highly modified as to render the
possible expression of state 1 very difficult.

Anterior apophysis present (Figs 422—425) (0)/absent (Figs 426, 429) (1).

Ventral margin of antrum not evaginated and projecting beyond the posterior edge of intersegmental
membrane A7—A8 (Fig. 439) (0)/ventral margin of antrum produced as an evaginated spinose or
heavily sclerotized projection over the ostium bursae (Figs 427, 428, 435) (1). This probably
corresponds to the ‘classic’ plusiine character of ‘ostium bursae unprotected/protected’. Many,
probably convergently derived, conditions are included under the apomorphic state. In Allagrapha,
Diachrysia and Brachionycha, the evagination is poorly sclerotized and membranous. In contrast, in
Palaeographa and Caloplusia, the evagination is very strongly sclerotized and the intersegmental
membrane is fused to the projection ventrally. Autoplusia is coded 0 as A. egena exhibits the
plesiomorphic state. However, ‘Syngrapha’ gammoides has a very weakly developed apomorphic
condition, similar to that found in Antoculeora. The state in Loboplusia is suspected to be 0 but it is
difficult to say as most of the membrane anterior to the lip of the antrum has been removed in the
preparation of the slide. In Puriplusia and Macdunnoughia, the evagination is limited to the
subventral area.

Lateral edges of antrum unmodified (0)/produced as two + spinose pads lateral to the ostium bursae
(Fig. 431) (1)/further fused with similar projections from the dorso-lateral corners of the antrum to
form recurved, spinose grooves (Fig. 432) (2). Many of the genera with broad antra (196:1), especially
among the outgroups, have somewhat similar lateral grooves to those found in Puriplusia and
Macdunnoughia but are not heavily spinose or as deeply invaginated.

Ventral margin of antrum and intersegmental membrane A7—A8 ventrally with short, dense, upright
scales (Fig. 433) (1)/without (0). The entire female genitalia in Diloba are surrounded by a dense pad
of upright, black scales. These, however, occur primarily on T7 and St7 and are considered
non-homologous to the upright antrum-mouth scales found in Allagrapha and other genera.

194 *Antrum without setae (0)/antrum with 1-3 setae subventrally or laterally (Fig. 435) (1)/antrum with

several setae along inner, ventral edge (Fig. 434) (2). Brachionycha has numerous setae lateral and
subventral to the antrum but, as these are a continuation of the generally setose condition of T8 and

110

195

196

197

198

199

200

201

202

IAN J. KITCHING

nota discrete patch, Brachionycha was coded 0. The setae along the inner, ventral edge of the antrum
in Diloba also form a continuation of those found along the posterior margin of A8 but the genus was
coded 2 because of their localized occurrence. However, it remains possible that the condition found
in Diloba represents a convergent development from the condition found in Brachionycha rather
than a homologue of that found in Pseudeva and Plusidia.

Anterior edge of T8 + level with ostium bursae (0)/T8 ventrally produced anteriorly to lie along the
antrum dorsally and exceeding it (Figs 426, 429) (1). This extension of T8 antero-ventrally may
represent a highly modified anterior apophysis (cf. character 190). This is suggested by the extension
bearing an apical, dorsal small lobe. However, Anuga has a lateral anterior bump on the margin of T8
in the position often occupied by the anterior apophysis in other genera. Therefore, until the
Stictopterinae and Euteliinae are examined in more detail, the present interpretation will be adhered
to. In Lophoptera, the extensions are apically highly modified into a pair of spinose pads. See also
character 202.

Antrum a distinctly invaginated, often heavily sclerotized, structure (Fig. 439) (0)/antrum a broad,
shallow, open bowl, ostium bursae somewhat exposed (Fig. 434) (1). Brachionycha and Stictoptera
have narrow, shallow antra and were coded 0. Diachrysia, Allagrapha, Puriplusia and Macdun-
noughia have invaginated but still somewhat open antra with marginal folds (see character 191) that
protect the ostium bursae. These genera were coded 0.

Antrum at least with sclerotized granulations, often very heavily sclerotized, well differentiated from
ductus bursae (0)/antrum membranous, barely differentiated from the ductus bursae (1).

*Antrum sclerotization (if present) free from T8 or continuous only at the antero-ventral corner (Fig.

434) (0)/antrum with two subdorsal, apically broad and truncate bars, T8 ventral edges extended
towards the midline (Fig. 433) (1)/extended T8 and antrum subdorsal bars fused, membranous area
between the ventral edges of T8 very narrow (Fig. 437) (2)/ventral edges of T8 fused; T8, LPV and
antrum forming a single, sclerotized area (Fig. 428) (3). Mouralia shows a state similar to state 2 but
the T8 edges are not parallel and have not been extended (as indicated by the presence, in this genus
of setae along the inner, ventral edge of T8). Setae are absent along the ventral edge of T8 in taxa
showing states 1—3 (but, it should be noted, such setae are also absent in several genera showing state
0; e.g. Omorphina, Diachrysia, Allagrapha, Syngrapha). Pseudoplusia has extended edges to T8 but
no subdorsal bars from the antrum. Autoplusia is coded 2 (the state shown by both A. egena and A.
olivacea) but ‘Syngrapha’ gammoides displays state 3.

Anterior apophysis arising approximately halfway down T8 in lateral view (Fig. 430) (0)/anterior
apophysis arising approximately level with the ventral edge of T8 (Fig. 440) (1). Taxa lacking apodeme
8 (character state 190:1) were coded *.

Antrum large, heavily sclerotized, smooth, open and curved somewhat to the RHS (Figs 437, 439)
(1)/antrum relatively small, often poorly sclerotized, with longitudinal folds and ridges, not forming an
open cavity (0). The antra of Eosphoropteryx and Palaeographa are curved to the LHS.

Membrane lateral and anterior to ostium bursae thin and unmodified (0)/membrane thick, white,
folded and tough (1). The thickened membrane in Allagrapha is the same membrane as that coded in
character 176 and is thus most probably convergent on that condition found in, for example,
Palaeographa. The membrane in Euchalcia is only slightly thickened.

T8 dorsally with anterior edge + level with or anterior to the antrum (0)/T8 represented by a thin band
dorsally, anterior edge posterior to the antrum (Fig. 426) (1). T8 in Adeva (Fig. 430) is reduced to a
thin band dorsally but this is formed by a strongly cut-back edge from an anteriorly positioned
anterior apophysis. It is considered to be an independent derivation from state 0 rather than
homologous to the condition found in the euteliines and Lophoptera.

Ductus bursae & corpus bursae

203

204

205

206

207

Ductus bursae (DB) short (less than half the length of abdomen), usually broad (Fig. 458) (0)/DB long
(over three-quarters the length of the abdomen) and narrow (Fig. 446) (1)/DB over 2X as long as
abdomen and very narrow (Fig. 447) (2).

Long axis of corpus bursae (CB) + parallel with long axis of DB (Fig. 449) (0)/long axis of CB + at
right angles to long axis of DB (Figs 446, 447, 448) (1).

Membrane of CB, at point of entry of DB, unmodified (0)/thickened, white and folded (1).

Median section of DB tubular or merely flattened (Fig. 446) (0)/median section expanded somewhat
laterally with dorsally recurved edges (Fig. 441) (1). Dactyloplusia has a laterally expanded median
section of the DB but the edges are not recurved.

DB with a lateral LHS short pouch with thick walls (1)/without (0).

208

SYSTEMATICS OF THE PLUSIINAE 111

CB relatively inelastic, + maximum size when dissected (0)/CB with very elastic walls, very small when
first dissected but can be very easily stretched to several times its original size (1).

209 *Ductus seminalis arising from apex of CB (Figs 450, 461, 462, 463) (0)/from fundus (Fig. 444) (1)/from

an accessory bursa produced from the DB (Fig. 460) (2). There is considerable difficulty in some taxa
(e.g. Mouralia, Diachrysia, Puriplusia) in deciding which of two pouches is an enlarged apex and
which is the CB proper. In Puriplusia, the orientation of the grooves near the origin of the DB
(character 213) suggests that there may have been a 180 degree change in the orientation of the CB so
that the apex is now directed anteriorly. However, this is not certain and pending further work,
Puriplusia is coded 1. The same situation may pertain to Loboplusia. In Sclerogenia, the apex is
interpreted to be recurved and totally fused with and open to the CB. Erythroplusia (Fig. 453) is also
very problematical. A fundular origin of the ductus seminalis is inferred but it could be that the CB is
extremely reduced and the apex very elongate to give the observed ‘boomerang-shape’, resulting in
the ductus seminalis then being interpreted as arising from the apex and changing the coding for the
genus to 0. The second species of Erythroplusia, E. pyropia, is very similar to E. rutulifrons and is of
no use in resolving this problem. Those genera exhibiting state 1 for character 207 also present
difficulties. The ductus seminalis is interpreted as arising from the apex in Trichoplusia and
Thysanoplusia, and from the fundus in Zonoplusia and Eutheiaplusia, but this is by no means certain.
Plusiopalpa is interpreted as having a small, recurved apex, bearing the ductus seminalis, with a
secondary large apical pouch present dorsal and posterior to it. The ductus seminalis in Pseudoplusia
is considered to arise from a long, anteriorly-recurved apex that is slightly fused to the CB basally (see
character 214). Chrysodeixis is interpreted as having a fundular origin but examination of other
species in this genus (C. acuta, [Dufay, 1970a]; C. illuminata, BM noctuid slide 11649) shows a similar
condition to that found in Pseudoplusia, but in which the origin of the recurved apex has moved to a
position about halfway along the CB. The extreme fundular origin of the ductus seminalis in C.
chalcites may merely represent the ultimate expression of this transformation. The origin of the CB
from the DB in Stictoptera is less well defined than in most species but the ductus seminalis still
appears to arise from the apex of the CB, rather than a lateral origin on the DB. A similar
interpretation was followed for Magusa, Paectes and Anuga, although the condition found in
Lophoptera casts some doubt on this interpretation. More work is evidently needed in the
stictopterine-euteliine complex.

210 *CB with weakly defined and irregular folds and lines, often smooth (Fig. 463) (0)/with regular,

211
Zl2

elongate, longitudinal ridges, originating from dorsal and lateral longitudinal ‘focal’ ridges, which may
or may not be sclerotized (Fig. 464) (1)/ridges forming a + circular pattern (Fig. 457) (2).

CB without a localized signum (0)/CB with a small, slightly concave, circular signum (Fig. 457)
(1)/signum invaginated as a narrow internal ridge (Fig. 465) (2).

CB with uniform internal ornamentation or, if several forms present, then these gradually merge one
into another (0)/CB with a distinct discontinuity halfway along its length, separating a pale, membra-
nous posterior half from an anterior half covered in sclerotized granulations and finer folding (Figs
454, 456) (1).

213 *Area of entry of DB into CB with large, strong, well sclerotized internal folds that extend onto the apex

214
215

or rarely onto the fundus (Figs 450, 451, 454, 455, 456) (3)/ridges present and sclerotized but not
strongly developed (Figs 445, 449) (2)/ridges not sclerotized, merely granular (Fig. 452) (1)/ridges
absent (Fig. 461) (0). Calophasia has a smooth area of sclerotization in the dorsal angle between the
DB and the apex, that may have a few folds. However, these folds are not extensive and the genus
was coded 0. Paectes has a long series of diagonal, heavily sclerotized folds on the posterior section of
the CB but these are not connected with folds arising on the DB and are not considered homologous
to the folds found in Autoplusia and related genera. The apex and folds are highly reduced in
Allagrapha, while in Puriplusia, they continue onto the fundus rather than the apex (but see 209).
Apex recurved in dorso-ventral plane and somewhat fused basally to the CB (Figs 443, 444) (1)/apex
not so modified (0).

CB without a dorso-lateral RHS patch of thickened white membrane somewhat anterior of the
entry-point of the DB (0)/such a patch present (Figs 461, 462) (1)/patch expanded into a shallow pouch
(Fig. 463) (2)/expanded further into a large pouch or a long appendix in which the thickened membrane
is present only apically (Fig. 458) (3).

| Miscellaneous

| 216

Colour pattern monomorphic (0)/pattern sexually dimorphic, males greyish, females brown (Figs 82,
83) (1). In Autoplusia, only ‘Syngrapha’ gammoides is obviously sexually dimorphic in this manner.
The difference is only barely discernible in other species of the genus (A. egena, A. olivacea). In

HSV IAN J. KITCHING

Sclerogenia, the females appear to be slightly duller than the males but this is not great enough to
warrant the description ‘sexually dimorphic’. Cucullia is also sexually dimorphic but shows a
different type of variation (males have white hindwings, females brown) and was coded 0. Stictoptera
is highly polymorphic but the various forms are not sex limited.

Analytical procedure

The data were analysed using David Swofford’s program for inferring phylogenies using the principle of
maximum parsimony, PAUP (version 2.2), run on the IBM System/370 3032 at the Rutherford Appleton
Laboratory. Due to the large size of the data set (57 taxa by 216 characters), the analysis was broken down
into several subprocedures that were run separately at different times.

Initially, a simple Wagner tree was constructed using the addition sequence CLOSEST and setting
HOLD (the number of trees held in memory at any one step of tree construction) equal to one. No
branch-swapping was employed at this point. The result was a single cladogram of 659 steps and a
consistency index (Kluge & Farris, 1969) of 0-466.

This cladogram was subsequently used as an input to a branch-swapping analysis (using the option
SWAP = ALT in an attempt to reduce the run time), which also invoked the MULPARS option to check
for the existence of multiple, equally-parsimonious cladograms. A default maximum of 50 cladograms was
found, with the reduced length of 648 steps and a consistency index of 0-474.

These 50 cladograms were then evaluated in order to determine the topological differences between
them. It was found that they consisted of five pairs, one triplet and one quadruplet of mutually-exclusive
subcladograms and a single alternative that was mutually exclusive to all but one of the five pairs.
Combining these alternative subcladograms gives a minimum number of 448 equally-parsimonious
cladograms for this data set.

In order to determine which of these 448 possibilities represented the best hypothesis of the interrela-
tionships of the plusiine genera, the apomorphy lists for six (output cladogram numbers 27, 29, 39, 44, 45
and 46) were obtained, which covered all the alternative subtopologies. A single preferred cladogram was
then identified by a process of character analysis and evaluation, which is described in the next section.

A second analysis was then performed on a restricted data set, with the outgroup taxa (Chalcopasta-
Anuga, 48-58) excluded. The intention was to check whether the 64 alternative topologies found for the
plusiines alone were exhaustive or whether, by restricting MAXTREE (the maximum number of
equally-parsimonious trees held in memory during a branch-swapping procedure) to 50, any had been
missed. To save computing time, the preferred topology (see below) was entered as the initial user-tree and
branch-swapping performed using SWAP = ALT, MULPARS and MAXTREE = 50. The initial
cladogram of 451 steps (it is worth noting in passing that 197 or 30-6% of the steps on the original cladogram
occurred outside the Plusiinae) was, surprisingly, reduced in length by two steps to 449 (consistency index
not calculated) and only 16 alternatives were found. The two missing pairs were those involving the
location of Loboplusia and the sequence of branching among the various tribes and subtribes (see below).

Initially, it was suspected that this additional shortening of the cladogram was an artefact caused by the
omission of the outgroup taxa. Thus, a final analysis was undertaken in which one of the previously
obtained outgroup topologies was added to one of the 16 plusiine topologies and the resultant cladogram
entered as a user-tree.

Fitting of the data to this cladogram resulted in a length of 647, a reduction of one step. Thus the 50(+)
cladograms of 648 steps obtained originally did not represent the minimum-length as was first thought. The
branch-swapping algorithm applied in the first analysis had failed to find the minimum-length topology
before the MAXTREE limit of 50 had been reached. Further, it is unlikely that it would have been found
even had MAXTREE been set to the system maximum of 150, given a minimum of 448 possible
cladograms of 648 steps and the basal position of the swap necessary to locate the shorter cladograms of 647
steps. (The algorithm appears to perform its branch-swapping starting with pairs of terminal taxa and
proceeding towards the base of the cladogram. Also, to find the 647-step cladograms, a swap involving the
rearrangement of subcladograms of 8 or more taxa was required, based upon an initial topology that
occurred only five times out of the original 50 cladograms and which was first found at cladogram 33.) Had
the second analysis of the restricted data set not been performed, the existence of the 16 cladograms of 647
steps would never have been revealed.

However, examination and evaluation of the apomorphy lists of the 647-step cladograms demonstrated
that the branching sequence of the major plusiine lineages was determined by very similar character state
changes to those present on one subset of the rejected 648-step topologies. Thus, the minimum-length
cladograms were rejected in favour of the 648-step, preferred cladogram referred to above. The full
reasoning behind this decision is given below.

SYSTEMATICS OF THE PLUSIINAE 113

The 648-step cladograms

1. Agrapha and Plusiopalpa

Two subcladograms were found in which the positions of Agrapha and Plusiopalpa differed. Subcladogram
1 (Fig. 9) places Agrapha as the sister-group of Argyrogramma + Stigmoplusia (A+S), and Plusiopalpa as
the sister-group of Acanthoplusia + Ctenoplusia (A+C). Subcladogram 2 (Fig. 8), in contrast, treats
Plusiopalpa as the sister-group of A+S, and Agrapha as the sister-group of these three genera plus A+C.
The different topologies are due to alternative interpretations of 13 characters (61, 79, 83, 86, 87, 88, 92,
98, 105, 130, 138, 144, 162). An evaluation of these characters is required in order to decide which
subcladogram represents the best cladistic hypothesis of the relationships between the six genera.

61:

79:

83:

105

130:
138:

144:

The sequence of suppression of the male fore-tibial spines forms a clear, non-homoplasious
transformation series on subcladogram 2, where state 1 characterizes A+S + Plusiopalpa, state 2
unites A+S, and state 3 is autapomorphic for Argyrogramma. Character 61 is not as convincing on
subcladogram 1, on which states 1 and 2 are interpreted as independent derivations from state 0. On
both cladograms, state 1 is also convergently derived in Plusiotricha and in the stictopterines (see
also below). Thus character 61 supports subcladogram 2.

This character is problematical on both subcladograms. The expected transformation series, in
which the hair tufts on segments A5 and A6 both gain discrete edges (state 1), followed by
independent losses of such margins on A6 (state 2) and AS5 (state 3), did not appear. On
subcladogram 1, A+S + Agrapha are characterized by the change 0 — 3 (discrete edges on A6 only),
followed by independent derivations of state 2 in Stigmoplusia (A6 only > AS only) and state | in
Agrapha (A6 only > AS and A6). On subcladogram 2, state 1 in Agrapha is independently derived
from state 0, with A+S characterized by the change 0 — 3 followed by the autapomorphic change
3 — 2 in Stigmoplusia. Alternative optimizations on subcladogram 2 are equally ambiguous. One
assigns state 0 to the node immediately ancestral to Argyrogramma and Stigmoplusia, in which case
states 2 and 3 respectively are independently derived in the two genera from state 0. Alternatively,
the internal node could be assigned state 2, whence Argyrogramma is characterized by the
autapomorphic change 2 = 3. Subcladogram 1 is marginally favoured in that character 79 delimits a
monophyletic group on this subcladogram but the transformation series is not very satisfactory,
despite a character consistency of 1-000 (true also of subcladogram 2).

This character is one of the most highly homoplasious in the data (consistency = 0-313). However,
state 4 is only found in Plusiopalpa and Argyrogramma and thus lends weak support to subclado-
gram 2 (Stigmoplusia possess the widely distributed state 3).

Subventral accessory pockets unite A+S and Agrapha as a monophyletic unit and thus support
subcladogram 1. On subcladogram 2, these pockets have to be hypothesised to have developed
independently in the two groups (as well as in Acanthoplusia, which is also true on subcladogram 1).
Although optimization of this character on subcladogram 2 suggests an initial gain followed by a
‘monophyletic’ loss in A+S + Plusiopalpa, an equally parsimonious interpretation is one of
independent gain in A+C and Agrapha. This is also the optimized distribution of states for this
character on subcladogram 1 and thus character 87 is of no use in deciding which subcladogram
should be accepted.

These characters show similar distributions to the last, in that they can be interpreted on subclado-
gram 1 as having been independently gained (Agrapha and Argyrogramma) or gained once (thus
characterizing A+S + Agrapha) with subsequent reversal (in Stigmoplusia). The former distribu-
tion is that found on subcladogram 2 and so characters 88, 98 and 105 are also irrelevant to the
decision.

On subcladogram 1, the state changes 1 — 3 and 1 — 2 arise independently and characterize A+ S
and A+C + Plusiopalpa respectively. On subcladogram 2, state 3 transforms into state 2, resulting
in the exclusion of Agrapha. Although both subcladograms are equally supported, other shared
similarities of the secondary hair pencil attachments between A+S and Plusiopalpa (see below) tip
the balance slightly in favour of subcladogram 2 and its sequential transformation series.

A juxta that is free from the saccular flanges also occurs in Pseudoplusia + Chrysodeixis and thus
offers only weak support for the monophyly of A+S + Agrapha (subcladogram 1).

Similarly, but conversely, the monophyly of A+S + Plusiopalpa (subcladogram 2) can be weakly
supported by the monophyletic reversal of this character.

This character (the degree of development of the clavus) shows an extremely high amount of
homoplasy (character consistency index = 0-133) and thus is very suspect. Both subcladograms
postulate the reduced condition (state 1) to be plesiomorphic for the six genera. Subcladogram 1
hypothesises independent reversals to a long clavus (state 0) in Argyrogramma and A+C +

114

IAN J. KITCHING

Argyrogramma Argyrogramma
Stigmoplusia Stigmoplusia
Plusiopalpa Agrapha
Acanthoplusia Plusiopalpa
Ctenoplusia Acanthoplusia
Agrapha Ctenoplusia
8 Plusiotricha 9 Plusiotricha
Eutheiaplusia Eutheiaplusia
Lamprotes Lamprotes
Pseudeva Pseudeva
Plusidia Plusidia
Euchalcia Euchalcia
Adeva Adeva
10 Panchrysia 11 Panchrysia
Chrysanympha Chrysanympha

Figs 8-11 8, subcladogram 2, depicting Plusiopalpa as the sister-group of Argyrogramma + Stigmoplusia
and a relatively basal position for Agrapha. 9, subcladogram 1, illustrating the alternative placements of
Plusiopalpa and Agrapha. 10, subcladogram 3, showing Plusidia as the sister-group of Lamprotes +
Pseudeva. 11, subcladogram 4, the alternative position of Plusidia as the sister-group of Euchalcia +
Adeva.

SYSTEMATICS OF THE PLUSIINAE Li)

Plusiopalpa, while subcladogram 2 suggests a reversal above the point where Agrapha branches off,
followed by a re-reversal to state 1 in Stigmoplusia. Both subcladograms include an autapomorphic
extreme reduction in Agrapha. Given the number of steps associated with this character, it is
considered to be too unreliable and of little use, especially with regard to the current decision.

162: A+S + Agrapha are united on subcladogram 1 by the (highly homoplasious) gain of a subapical
cornutus (SAC). However, examination of species other than the type species shows a SAC to be
general in Ctenoplusia (Dufay, 1970a). C. limbirena is apparently aberrant in lacking one, so further
increasing the homoplasy in this character. Thus, absence of a SAC is rejected as demonstrating
relationship.

Overall, the decision between the two topologies was close. Subcladogram | is supported by character 86
and weakly by characters 79 and 130. Subcladogram 2 is supported by character 61 and weakly by
characters 83, 92 and 138. On balance, the evidence in favour of subcladogram 2 appears slightly better and
it is accepted over subcladogram 1. However, the most expedient solution may be the consensus
tetrachotomy (see below).

2. Plusidia

Two subcladograms were found in which the position of Plusidia was ambiguous. In subcladogram 3 (Fig.
10), Plusidia is treated as the sister-group of Lamprotes + Pseudeva (L+P), while in subcladogram 4 (Fig.
11), the genus is placed as the sister-group of Euchalcia + Adeva(E+A). Ten characters (6, 41, 42,58, 144,
154, 171, 186, 194, 205) are involved in the ambiguity.

Five characters support a sister-group relationship between Plusidia and E+A. Of these, four are
reversals in highly homoplasious characters (6, 41, 42, 58), while the apomorphic state for the fifth (205)
also characterizes Cucullia, Trichoplusia and Plusiotricha + Eutheiaplusia. The evidence is thus not very
strong. In contrast, the five characters that suggest that Plusidia is most closely related to L+ P include two
(171 and 186) that are unique and unreversed. Further, the transformation series of character 194 is only
homoplasious insofar as state 1 is independently derived in the outgroup taxon Diloba, and there is doubt
as to the homology of this condition and that found in Plusidia + L+P (see below). Character 154 is found
in its apomorphic state in Panchrysia, Chrysanympha and E+A and thus is interpreted as reversed in

_ Plusidia + L+P on subcladogram 3, while the last, character 144, is one of those for which extreme doubt
was expressed in the previous section regarding reliability.

Overall, there can be little doubt, after evaluating the characters supporting the alternative subclado-
grams, that subcladogram 3 is to be preferred and Plusidia interpreted as the sister-group of Lamprotes +
Pseudeva.

3. Autographa, Cornutiplusia and Plusia

It was noted during the data collection that few synapomorphies were being found for Autographa,
Cornutiplusia, Plusia and their immediate relatives. Thus, it was not unexpected when multiple, equally-
parsimonious subcladograms were found that included these genera. In all of these subcladograms, there
was a constant terminal cluster of four genera (Caloplusia, Syngrapha, Palaeographa, Lophoplusia; =
CSPL) and a constant outgroup, Anagrapha. Between these taxa, four topologies were found that
rearranged Plusia, Cornutiplusia, Autographa gamma and A. bractea.

Three subcladograms, 5, 6 and 7 (Figs 12—14 respectively) agree in placing A. gamma as the basal taxon,
while subcladogram 8 (Fig. 15) so places A. bractea. Three subcladograms, 5, 7 and 8, also agree in placing
_ Plusia as the sister-group of CSPL, while subcladogram 6 places Cornutiplusia in this position. Subclado-
gram 5 differs from subcladogram 7 in placing Cornutiplusia and A. bractea as sister taxa. Four characters
only are involved in differentiating the alternative topologies: 12, 64, 166 and 215.

Character 12 is highly homoplasious (consistency index = 0-100). The apomorphic state varies from a
mere silver spot between the reniform stigma and the cubital vein (as seen in Antoculeora and Plusiopalpa)
to that in which the reniform stigma is almost entirely outlined in silver (a condition unique to Lophoplusia
and perhaps not too far removed from the completely silver stigmata of Plusia and A. bractea). However,
character 12 is one of the two characters that support a sister-group relationship on subcladogram 6
between Cornutiplusia and CSPL. The other character on this branch, character 64, is the ‘classic’ plusiine
character of spined hind tibiae. This is also highly homoplasious; state 1 is also found in Autoplusia and
state 2in Rachiplusia and Anagrapha. The topology of subcladogram 6 is also supported by the observation
that occasional spines occur in Plusia and Autographa (McDunnough, 1944).

Plusia is placed as the sister-group to CSPL on subcladograms 5, 7 and 8 by a synapomorphy in character
166. State 2, the distinctly winged subapical cornutus, was scored as autapomorphic for Lophoplusia and as
‘not comparable’ for Palaeographa, Syngrapha and Caloplusia, because the type species of these taxa lack
subapical cornuti. That these three taxa are predicted to show state 2 is interesting. Several species of

116

12

14

IAN J. KITCHING

Caloplusia
Syngrapha
Palaeographa
Lophoplusia

Plusia

Autographa bractea

Cornutiplusia

Autographa gamma

Anagrapha

Caloplusia
Syngrapha
Palaeographa
Lophoplusia

Plusia

Cornutiplusia

Autographa bractea

Autographa gamma

Anagrapha

13

15

Caloplusia
Syngrapha
Palaeographa
Lophoplusia

Cornutiplusia

Plusia

Autographa bractea

Autographa gamma

Anagrapha

Caloplusia
Syngrapha
Palaeographa
Lophoplusia

Plusia

Cornutiplusia

Autographa gamma

Autographa bractea

Anagrapha

Figs 12-15 The four alternative subtopologies involving Autographa, Plusia and Cornutiplusia. 12,
subcladogram 5. 13, subcladogram 6. 14, subcladogram 7. 15, subcladogram 8.

SYSTEMATICS OF THE PLUSIINAE 117

Syngrapha, particularly those considered by Eichlin & Cunningham (1978) to be primitive (e.g. S.
ottolenguii (Dyar), S. viridisigma (Grote), S. orophila (Hampson)), do bear subapical cornuti on the vesica
and in one species, S. epigaea (Grote), the cornutus even appears to be winged (Eichlin & Cunningham,
1978: fig. 58a). Thus, the interpretation of Plusia as the sister-group of CSPL, on the basis of character
state 166:1, may prove well-founded and is provisionally accepted here as the best cladistic hypothesis.
Subcladogram 6 is thus rejected, despite the evidence furnished by character 64.

A. gamma is placed as the most basal taxon on subcladograms 5—7 because it displays the plesiomorphic
state for character 215. The first apomorphic state is the presence of a dorso-lateral patch of thickened
membrane on the corpus bursae and is found in Plusia and Lophoplusia. This is further elaborated into a
shallow pouch (state 2) in A. bractea and into a large pouch or appendix (state 3) in Cornutiplusia. All
subcladograms require a reversal from state 1 to state 0 in Palaeographa + Syngrapha + Caloplusia.
Subcladogram 7 has what appears to be the worst optimization of states, requiring the derivation of state 2
directly from state 0, then a further autapomorphic development of state 3 in Cornutiplusia, together witha
reduction in the structure, stage 2 — 1, along the branch leading to Plusia + CSPL. Subcladogram 8 is
somewhat better in that independent derivations of states 2 and 3 from state 0 are postulated, a distribution
of states also found on the rejected subcladogram 6. Subcladogram 5, however, unites A. bractea and
Cornutiplusia using state 2 and then derives state 3 autapomorphically in the latter taxon. This interpreta-
tion is further supported by the observation that state 2 appears to be present in all the putative North
American relatives of A. bractea except one, as judged from the figures and phylogenetic tree given by
Eichlin & Cunningham (1978) (but see Clade analysis). Thus, the branching sequence, and associated
character state changes, of subcladogram 5 is accepted as the best cladistic hypothesis of relationships for
these genera.

4. Loboplusia

The genus Loboplusia, with its large amount of missing data and relatively simple male genitalia, was
expected to be ambiguous in position. Two subcladograms were found; subcladogram 9 (Fig. 16) places
Loboplusia as the sister-group of Erythroplusia, while subcladogram 10 (Fig. 17) hypothesises a sister-
group relationship between Loboplusia and Autoplusia + Antoculeora + Sclerogenia (AAS). Eleven
characters (6, 12, 59, 113, 142, 148, 155, 176, 177, 209) are involved in these alternative interpretations of
the position of Loboplusia.

Loboplusia is grouped with Erythroplusia on the basis of three character state changes. The loss of the
scale tooth on the forewing tornus (6) is highly homoplasious and of little importance. Similarly, the
fundular origin of the ductus seminalis on the corpus bursae (209) appears to have been multiply derived,
supporting the reservations outlined above regarding the correct interpretation of corpus bursae morphol-
ogy. Finally, character 155 was scored as ‘missing data’ for Loboplusia, because the transtillae were torn.
Thus, the presence of the apomorphic state of character 155 in Loboplusia is a prediction of the
optimization procedure and not an empirical observation. The evidence supporting subcladogram 9 is
therefore tenuous.

On subcladogram 10, Loboplusia is placed by four characters in a larger clade including AAS and a clade
consisting of Allagrapha, Macdunnoughia and Puriplusia (AMP). Two of these characters, 142 and 176,
are subsequently reversed on the branch subtending Loboplusia. The condition of character 177 was
considered somewhat uncertain but does provide weak support for this topology. The apomorphic state of
character 113 is shared by Loboplusia and AMP, and is considered to have reversed in AAS. Further,
Loboplusia is placed as the sister-group of AAS by three characters, all with low consistency indices. Two
(59 and 148) are reversals that occur several times elsewhere on the cladogram, while the third (12) was
rejected above as unreliable.

Overall, the evidence supporting each placement of Loboplusia is very weak, with subcladogram 10
being marginally preferable. However, until such time as further studies are performed and additional
specimens become available, the best course may be to treat Loboplusia as forming a trichotomy with
Erythroplusia and AAS + AMP (see below).

5. Rachiplusia and the branching sequence of the tribes and subtribes
The final pair of alternative topologies within the Plusiinae affects the fundamental branching pattern of
the Argyrogrammatini and the three subtribes of the Plusiini: the Plusiina, Euchalciina and Autoplusiina.
(The Autoplusiina are temporarily defined as Rachiplusia, Erythroplusia, Loboplusia, AAS and AMP.) In
addition, there is ambiguity concerning the relationships of Rachiplusia.

Subcladogram 11 (Fig. 18) agrees well with the trees put forward previously by Ichinosé (1962b) (Fig. 2)
and Eichlin & Cunningham (1978) (Fig. 3). Both the Argyrogrammatini and Plusiini, as currently
recognized, are monophyletic. Within the latter, the subtribe Euchalciina is the sister-group of the Plusiina

118 IAN J. KITCHING

—

Autoplusia Autoplusia j
Antoculeora Antoculeora

Sclerogenia Sclerogenia

Allagrapha Loboplusia
Macdunnoughia Allagrapha

Puriplusia Macdunnoughia
Loboplusia Puriplusia
Erythroplusia Erythroplusia

16 Rachiplusia 17 Rachiplusia

PLUSIINA ARGYROGRAMMATINI
AUTOPLUSTINA Rachiplusia
Rachiplusia AUTOPLUSIINA
|
EUCHALCIINA PLUSIINA
ARGYROGRAMMATINI EUCHALCIINA
18 ABROSTOLINI 19 ABROSTOLINI

Figs 16-19 16, subcladogram 9, showing Loboplusia as the sister-group of Erythroplusia. 17, subclado-
gram 10, with Loboplusia as the sister-group of Sclerogenia + Antoculeora + Autoplusia. 18, subclado-
gram 11, the topology of the tribal and subtribal branching that places Rachiplusia as the sister-group of
the remaining Autoplusiina. 19, subcladogram 12, the alternative subtribal topology illustrating a
paraphyletic Plusiini and with Rachiplusia as the sister-group of the Argyrogrammatini.

SYSTEMATICS OF THE PLUSIINAE 119

+ Autoplusiina. The genus Rachiplusia is placed as the most basal branch within the last of these subtribes.

The monophyly of the Plusiini is based upon three characters. Two of these (121 and 127) are moderately
homoplasious but the last, character 117, is uniquely derived but reversed in Antoculeora. In this genus, the
broad lateral flange of the pleurite is absent, a condition probably correlated with the autapomorphic form
of the pleurite coded by character state 118:2. The sister-group relationship between the Plusiina and the
Autoplusiina is based upon a single character, 116, the autapomorphic state of which is lost in Antoculeora
(cf. character 117) and independently derived in Anadevidia. Rachiplusia is placed as the sister-group of
the rest of the Autoplusiina based on reversals in three highly homoplasious characters (18, 59 and 148) and
a change from state 0 to state 2 in character 213. Although this character is homoplasious (particularly
when the outgroups are taken into consideration), further work may justify it and the position of
Rachiplusia it supports.

The alternative subcladogram 12 (Fig. 19) is in considerable disagreement, in that the Plusiini is
paraphyletic, not only with respect to the subtribal arrangement, but also because Rachiplusia is treated as
the sister-group of the Argyrogrammatini. However, the supporting characters are very weak. The
Argyrogrammatini, Autoplusiina and Plusiina are grouped together to the exclusion of the Euchalciina by
two characters. The first, 116, was synapomorphic for the Autoplusiina and Plusiina on subcladogram 11,
on which the presence of state 1 in Anadevidia (the basalmost member of the Argyrogrammatini) was
interpreted as convergent. On subcladogram 12, state | is interpreted as uniquely derived but reversed in
the Argyrogrammatini above Anadevidia. The second character state change is 83:0 — 3 and concerns the
degree of development of the sternal hair pencils in the males. The sequence of state changes for character
83 is very complex and highly homoplasious. Very little confidence can be placed in it. The Autoplusiina
and Argyrogrammatini are united by two reversals in characters 18 and 148, both of which are subject to
multiple convergent gains and reversals. Finally, two character state changes support the sister-group
relationship between the Argyrogrammatini and Rachiplusia: 14:0 — 2 and 110. The latter (whether the
vinculum is V-shaped or Y-shaped) is trivial, homoplasious and can be discounted. 14 can also be rejected
once its full transformation on subcladogram 12 is analysed. The complete series consists of 0 — 2
(Rachiplusia + Argyrogrammatini), 2 — 1 (Anadevidia only) and two convergent 0 — 2 changes
(Autoplusia, Mouralia). The first two of these imply that seta 1 of the female frenulum was first lost and
then partially regained as a rudiment in Anadevidia. The more parsimonious sequence morphologically, in
which the seta is reduced and then lost, is precluded by the position of Rachiplusia. However, both
interpretations for the origin of state 1 in Anadevidia (reduction as an intermediate step or as a terminal
reversal) are allowed on subcladogram 11. Thus, on the basis of this, and on the evidence provided by
characters 116, 117 and 213, subcladogram 12 is rejected and subcladogram 11 accepted as the best cladistic
hypothesis for this section of the plusiine cladogram.

6. Outgroups

Due to the variety and ‘taxonomic distance’ between the outgroups (11 genera from four subfamilies), it
was not surprising to find that there were many alternative topologies for them. Three constant features
were found:

(a) Oncocnemis and Chalcopasta are sister-groups;

(b) Brachionycha and Diloba are sister-groups; and

(c) the Stictopterinae and Euteliinae are each monophyletic and together form a monophyletic group to
which Magusa is the outgroup (= SEM).

However, the sequence in which these three groups and the other two genera, Cucullia and Calophasia,
clustered proved very variable. One pair of alternative topologies was found in which Calophasia +
Cucullia + SEM formed a clade. The first (Fig. 20) places Calophasia as the sister-group of SEM; the
second (Fig. 21) places it as the sister-group of Cucullia. Accepting this as a terminal taxon, three further
topologies were found in which its sister-group was Brachionycha + Diloba (Fig. 22), Oncocnemis +
Chalcopasta (Fig. 23), or these four genera together (Fig. 24). In addition, a non-exclusive alternative

topology was found (Fig. 25) in which Calophasia + Cucullia was the sister-group of Oncocnemis +
Chalcopasta, and SEM was the most basal clade. All these relationships have to be qualified by the
statement that there is no guarantee that the root of the entire cladogram lies between the Plusiinae and the
outgroups.

The precise resolution of the relationships of the outgroups is of secondary importance in this study.
Thus, the decision as to which of the above alternatives is to be preferred is not critical. The topology that is
depicted on the final preferred cladogram (see below) is that which places the Plusiinae nearest to the
Stictopterinae, their erstwhile closest relatives, in order to evaluate the characters that support this
relationship.

120

IAN J. KITCHING

Stictopterinae Stictopterinae
Euteliinae Euteliinae
Magusa Magusa
Calophasia Calophasia
20 Cucullia 21 Cucullia
Stictopterinae Stictopterinae
Euteliinae Euteliinae
Magusa Magusa
Calophasia Calophasia
Cucullia Cucullia
Brachionycha Oncocnemis
Diloba Chalcopasta
Oncocnemis Brachionycha
22 Chalcopasta 23 Diloba
: : Stictopterinae
Stictopterinae a
ae Euteliinae
Euteliinae
Magusa
Magusa
Calophesia 5
B h h
eee a rachionycha
Diloba
Brachionycha
Oncocnemis
Diloba Chalcopasta
Oncocnemis Calophasia
24 25 Cucullia
Chalcopasta

Figs 20-25 The alternative outgroup subtopologies. 20, outgroup subtopology 1. 21, subtopology 2. 22,
subtopology 3. 23, subtopology 4. 24, subtopology 5. 25, subtopology 6. Note that subtopology 6 is an
exclusive alternative to subtopologies 1—5S.

SYSTEMATICS OF THE PLUSIINAE 121
The 647-step cladograms

Following removal of the outgroups, further global branch-swapping with the multiple parsimony option
discovered a set of 16 cladograms. After reinstatement of the outgroups, these cladograms were found to
be 647 steps long, one step shorter than those found in the earlier analyses that were described above. The
647-step cladograms are of similar topology to the rejected 648-step subcladogram 12, in that Rachiplusia
forms the sister-group of the Argyrogrammatini, but differ in the branching sequence of the subtribes of
the Plusiini (Fig. 26). Whereas, on subcladogram 12 of the 648-step cladograms, the Autoplusiina (less
Rachiplusia) form the sister-group of the Argyrogrammatini + Rachiplusia, on the 647-step cladograms, it
is the Plusiina that occupy this position. The 16 alternatives arise from the mutually exclusive topologies
referred to previously as subcladograms 1/2 (Agrapha/Plusiopalpa), 3/4 (Plusidia) and 5/6/7/8 (Auto-
grapha/Cornutiplusia/Plusia). The presence of further alternatives in the outgroup taxa was not observed
but all of the topologies found previously would be expected. Reversing the order of branching of the
Plusiina and Autoplusiina (less Rachiplusia) appears to preclude the alternative topology that places
Loboplusia as the sister-group to Erythroplusia.

ARGYROGRAMMAT INI

Rachiplusia

PLUSIINA

AUTOPLUSIINA

EUCHALCIINA

ABROSTOLINI

Fig. 26 The branching sequence of the tribes and subtribes on the 647-step cladograms. Note that the
position of Rachiplusia is the same as on subcladogram 12 of the 648-step cladograms (Fig. 19) but that
the Autoplusiina and the Plusiina have exchanged positions.

However, the 647-step cladograms (which, given that the branch-swapping procedure ran to comple-
tion, are most likely the shortest for this data set) are based upon the same characters (with one exception)
as the previously rejected 648-step subcladogram 12. Both topologies unite the Argyrogrammatini,
Plusiina and Autoplusiina using characters 83:0 — 3 and 116, and treat Rachiplusia as the sister-group of
the Argyrogrammatini based on characters 14:0 — 2 and 110. In addition, the latter relationship is
supported on the 647-step cladograms by reversals in characters 18 and 148, both of which are highly
homoplasious. Finally, the Plusiina are placed as sister-group to the Argyrogrammatini + Rachiplusia
solely on the basis of character 162. This character (presence/absence of a subapical cornutus) is also highly
homoplasious and was rejected as reliable in connection with the discussion of the relative positions of
Agrapha and Plusiopalpa. Thus, the 647-step minimum-length cladograms can no more be accepted than
those 648-step cladograms that include subcladogram 12. Therefore, they are rejected in favour of a longer
cladogram but one in which the supporting characters are more reliable.

Evaluation of the preferred cladogram

Following from the decisions reached in the previous two sections, the preferred cladogram (Figs 27, 28) is
648 steps in length and consists of subcladograms 2, 3, 5, 10 and 11; the outgroup topology that places the
Stictopterinae closest to the Plusiinae; and the sections held in common by all the alternative 648-step
cladograms. The first part of the evaluation of the (preferred) cladogram is character-oriented insofar as
the emphasis is placed primarily upon the branches along which a given character changes. Each character
will be taken in turn and its character state distribution on the cladogram discussed, with particular

122 IAN J. KITCHING

12

10,

Fig. 27 The preferred 648-step cladogram showing the branching sequence of the outgroups and the
Plusiinae except for the Plusiini (see Fig. 28). The numbers on the nodes are the clade reference numbers
used to identify clades in the evaluation of characters and taxa.

51

15,

47

48

52

13

11

49

50

53

54

55

PLUSIINI
Argyrogramma
Stigmoplusia
Plusiopalpa
Ctenoplusia
Acanthoplusia
Agrapha
Eutheiaplusia
Plusiotricha
Trichoplusia
Thysanoplusia
Zonoplusia
Dactyloplusia
Chrysodeixis
Pseudoplusia
Anadevidia
Abrostola
Mouralia
Omorphina
Magusa
Stictoptera
Lophoptera
Paectes

Anuga
Oncocnemis
Chalcopasta
Calophasia
Cucullia
Brachionycha

Diloba

19

SYSTEMATICS OF THE PLUSIINAE

45
44
43
42
40
39
38
37
4i
28
36
35
34
31
30
29
32
33
27
26
24
22
21
25
20
23

123

Caloplusia
Syngrapha
Palaeographa
Lophoplusia
Plusia
Cornutiplusia
Autographa bractea
Autographa gamma
Anagrapha
Diachrysia
Autoplusia
Antoculeora
Sclerogenia
Loboplusia
Allagrapha
Macdunnoughia
Puriplusia
Erythroplusia
Rachiplusia
Lamprotes
Pseudeva
Plusidia
Euchalcia
Adeva
Panchrysia
Chrysanympha
Polychrysia

Eosphoropteryx

Fig. 28 The preferred 648-step cladogram depicting the branching sequence of the genera of the tribe
Plusiini. The numbers are the clade reference numbers.

124 IAN J. KITCHING

reference to any homoplasy, alternative optimizations and possible errors in coding and scoring states. The
character consistency index (c) of each character is given in parentheses after the character number. The
second part of the evaluation is taxon-oriented and examines each of the clades and its supporting
characters in turn. The aim here is to determine to what degree each clade may be expected to be
corroborated in the light of further data. Clades without formal names are referred to by numbers that are
given on the cladograms in Figs 27, 28.

Character analysis

Wing pattern

1 (0-333) The presence of raised scales in the forewing cell is one of the classic Hampsonian noctuid
characters. Hampson (1913) noted their occurrence in the Plusiinae in Abrostola and Mouralia, and
also in Chrysanympha, the sole species of which he placed in Abrostola. Raised scales in Chrysany-
mpha were not observed by Eichlin & Cunningham (1978), nor were they in this study. This character
state also occurs elsewhere in the Noctuidae, most notably as a defining character for the Sarrothripi-
nae and as a secondary character for distinguishing the Stictopterinae from the Euteliinae. However,
in this study, the forewing of Lophoptera (Stictopterinae) was found to be smooth, while raised scales
were observed in Anuga (Euteliinae). As a result, 1 has a low consistency, being convergently
derived in the Abrostolini, Stictoptera and Anuga. It weakly supports the monophyly of the former
but more extensive sampling is required to adequately document its occurrence in the Stictopterinae
and Euteliinae.

2 (1-000) This character is apomorphic for the Abrostolini. Superficially similar patterns occur
elsewhere (e.g. Paectes abrostoloides (Guenée)) but lack this element.

3 (0.143) The Y-mark is one of the characteristic elements of the forewing pattern of the Plusiinae.
Within the subfamily, it appears to be synapomorphic for the Argyrogrammatini + Plusiini, but with
multiple reversals. In some instances (Anadevidia, Allagrapha), the lack of a Y-mark appears to be
part of a general simplification of the wing pattern, while in others (Thysanoplusia, clade 22), it is due
to further modification and elaboration of the pattern and possible overlaying of the Y-mark by more
derived elements. The latter hypothesis is supported in Thysanoplusia by the observation that only T.
intermixta and a few, apparently closely related, species lack the Y-mark (see character 9), while the
remaining species (e.g. T. indicator (Walker), T. daubei) possess a more typical pattern including a
Y-mark, albeit somewhat modified. Diachrysia (q.v.) merits further attention because in some
species (e.g. D. leonina (Oberthir)), the pattern is as simple as in Allagrapha, but in others (e.g. D.
chryson, and more so, D. chrysitis and D. zosimi (Hiibner)), this simple pattern 1s further elaborated
with large areas of golden-green metallic scales (see character 9). Within clade 22, Lamprotes
appears to have regained the Y-mark. This result suggests that the small golden C-mark in this genus
may not be homologous with the Y-mark.

An alternative optimization of character 3, in which the Y-mark is convergently gained in clade 6
(Argyrogrammatini less Anadevidia) and the Plusiini, with the lack in Anadevidia being treated as
plesiomorphic, is rejected as a less satisfactory interpretation.

4 (0.667) State 1 for this character is synapomorphic for the Argyrogrammatini and Plusiini, although
there are independent reversals in Zonoplusia and Eutheiaplusia. State 4, in which the pink line is
replaced by a white line, is independently derived from state 1 and is synapomorphic for Euchalcia
and Adeva. There are alternative optimizations for states 2 and 3, the best hypothesising the former
state to be a synapomorphy of Ctenoplusia and Acanthoplusia, with the latter condition being
autapomorphic for Ctenoplusia. However, state 3 is not universal within Crenoplusia, but may help
delimit a subset of this genus including not only C. limbirena, but perhaps also such species as C.
melanocephala (Méschler), C. micans (Dufay) and C. polycampta Dufay.

5 (0-200) Scattered, individual black scales on the forewing appear to be convergently derived in all
the genera that possess them. This was not unexpected for Erythroplusia, in which such scales are
present in one of the species (E. rutulifrons) but absent in the other (E. pyropia). In addition, this
genus is only distantly related to the other genera that possess these scales, most of which are
members of the Euchalciina. Within the Euchalciina, state 1 has been independently gained in
Polychrysia, Chrysanympha, Adeva and Pseudeva, suggesting that the correct homologies have not
yet been discovered. Further sampling, however, is unlikely to improve matters. A cursory
examination of the other species in the genera concerned showed that black scales were absent
entirely in all species of Panchrysia, Euchalcia and Plusidia (providing Plusidia imperatrix Draudt is
placed in Polychrysia, see below), and present in all Polychrysia and Pseudeva, although only weakly
developed in P. palligera (Grote). Scattered black forewing scales can also be seen in the unanalysed
Syngrapha parilis (Fig. 96).

6

SYSTEMATICS OF THE PLUSIINAE WS

(0-100) The presence or absence of the tornal scale tooth is a highly variable and homoplasious
character. Although characteristic of the Argyrogrammatini and Plusiini, it is multiply lost and
regained, with numerous alternative sequences of change. One such sequence in clade 4 postulates
acquisition followed by subsequent loss in Agrapha, Anadevidia, Erythroplusia, Loboplusia, clade
24 and clade 40, with regain in Palaeographa, Autographa bractea and Lamprotes + Pseudeva. The
scale tooth is evidently a highly unreliable character.

(0-500) Yellow hindwings occur scattered throughout the Plusiinae and seem to be partially
correlated with an arctic-alpine habitat. In the taxa under study, they have been independently
developed in Omorphina and Syngrapha + Caloplusia. Whilst yellow hindwings are characteristic of
the group of species related to S. devergens and C. hochenwarthi (e.g. C. tibetana Staudinger, S. ignea
(Grote), S. microgamma (Hiibner), S. montana (Packard), S. parilis [although this last species has
unique white hindwings]), they also appear to have been independently derived (Eichlin &
Cunningham, 1978) in another group of syngraphas, which includes S. diasema (Boisduval), S. borea
(Aurivillius), S. diversigna Ottolengui, S. orophila, S. sackenii (Grote) and S. snowii (Edwards).
Additional plusiines with yellow hindwings include Trichoplusia exquisita (Felder) (Fig. 45),
Panchrysia dives (Eversmann), Rachiplusia virgula (Blanchard) and Autographa sansoni Dod. This
colour pattern may prove to be of use within Syngrapha (s.1.) in a wider study but should be treated
with caution.

(1-000) A hyaline hindwing base is apomorphic for the Stictopterinae.

(0-333) Extensive patches of green-gold scales have been independently acquired by Chalcopasta,
Thysanoplusia and Diachrysia. The former supports Hogue’s (1963) contention that this colour
pattern is convergent in the stiriines and plusiines. This analysis also demonstrates that the pattern
similarity between Thysanoplusia and Diachrysia is also superficial.

(0-500) Similarly, a less pronounced, general metallic sheen on the forewings is also convergent
between Chalcopasta and the plusiines.

(0-333) The black spot at the apex of forewing vein M; is interpreted as convergent in Stigmoplusia,
Plusiotricha and Pseudoplusia + Chrysodeixis.

(0-100) This character, the degree of silver scaling around the reniform stigma, has already been
dismissed above, when choosing between subcladograms 5-8, as virtually useless for providing
evidence of relationship.

(0-500) The loss of the hair-like scales on the forewing cell underside is interpreted as convergent in
Chalcopasta and the euteliines + stictopterines.

Wing structure

14

15

16

17
18

19

(0-600) Three independent transformation series appear to be present in this character, all of which
treat the widespread state 0 as plesiomorphic and all of which have alternative sequences of state
change. The most parsimonious for each in terms of logical transformations is as follows. Within the
Plusiinae, the Argyrogrammatini are characterized by the change 0 — 1, while clade 6 is defined by
change 1 — 2. The change 0 — 2 has been convergently derived in Rachiplusia and Autoplusia. The
Stictopterinae + Euteliinae are united by the transformation 0 — 3. There are then two further
independent changes: 3 — 4 characterizes Paectes, while 3 5 is apomorphic for the Stictopterinae.
The node uniting Brachionycha and Diloba can take a value of 0, 2 or 6. An independent origin of
state 2 in Brachionycha and of state 6 in Diloba is supported by the presence of state 0 in
Brachionycha nubeculosa (Esper), Psaphida resumens Walker and Copivaleria grotei (Morrison).
(0-500) As was expected, the transformation series of reduction in hindwing vein M), (the trifine/
quadrifine dichotomy) did not emerge non-homoplasious. The plesiomorphic state remains uncer-
tain but the Plusiinae can be delimited by state | (although this is convergently derived in Diloba, and
Mouralia possesses state 2); the Stictopterinae + Euteliinae by state 0; and Calophasia by state 3.
Whatever the eventual outcome, it is certain that the trifine condition will be found to have been
attained on several, if not numerous, occasions.

(0-250) The apically-produced forewing at vein Ry, is interpreted as convergently derived in
Lamprotes + Pseudeva, Panchrysia, Allagrapha and Diachrysia.

(0-500) This venational character unites the Stictopterinae and Euteliinae but is reversed in Paectes.
(0-200) The origin of vein M, in the hindwing is partially correlated with character 15 and the
development of the trifine condition. It is even more highly homoplasious, with numerous converg-
ences and reversals, which include independent derivations of state 3 from both state 1 and state 2.
Many genera appeared intermediate between states 1 and 2 and this character is perhaps best
disregarded.

(0-200) The apomorphic state for the outline shape of the hindwing has been developed five times

126

IAN J. KITCHING

independently in Mouralia, Panchrysia, Magusa, Calophasia + Cucullia and Lamprotes + Pseudeva,
although the precise condition in Panchrysia and Lamprotes is unclear. Much of the homoplasy may
be due to its partial correlation with character 15.

20 (0-667) The plesiomorphic state for the scale arrangement appears likely to be state 1, with state 2
being a good synapomorphy for the Stictopterinae + Euteliinae + Magusa. State 0 is interpreted as
convergent in Omorphina and Brachionycha + Diloba.

21 (1-000) The presence of a short distal section of forewing vein CuP is synapomorphic for Bra-
chionycha and Diloba.

Head

22 (0-333) Lashed eyes characterize the Plusiinae and clade 51 of the outgroups, with reversals in
Brachionycha and Chalcopasta. Very long lashes have been independently acquired in Abrostola,
Cucullia and Diloba.

23 (1.000) Bulbous eyes unite the Stictopterinae, Euteliinae and Magusa.

24 (0-500) The prominent rim around the eyes has been convergently gained in Omorphina and
Syngrapha + Caloplusia. This character is an alternative statement of the ‘classic’ character ‘eye
small and elliptical’, and has been functionally correlated with a diurnal habit (Hardwick, 1970: 15;
see also Kitching, 1984: 184).

25 (1-000) A lower frons that is scaled only at the edge unites the Stictopterinae and Euteliinae.

26 (1-000) Like character 23, character 26, the degree of prominence of the frons, unites the Stictopter-
inae, Euteliinae and Magusa, and may only be an alternative formulation of that character.

27 (0-667) Triangular pilifers are apomorphic for the Plusiinae, although they are convergently derived
in Calophasia. Broad, reduced pilifers are autapomorphic for Diloba.

28 (1-000) A dorsally delimited frons unites the plusiine tribes Abrostolini, Argyrogrammatini and
Plusiini.

29 (1-000) A convex occiput is apomorphic for the Plusiinae.

30 (0-500) A narrow frons in the male is interpreted as convergent in Stictoptera and Anuga.

Proboscis

31 (1-000) A reduction in the number of apical sensilla on the proboscis characterizes the Plusiinae,
while an independent total loss unites Brachionycha and Diloba. This latter condition is probably
a direct correlate of the extreme reduction of the proboscis found in these two genera (character
32):

32 (0-667) As referred to above, the proboscis in Brachionycha and Diloba is very reduced and
non-functional. Most outgroup genera have a somewhat short, stout proboscis but in one pair,
Calophasia + Cucullia, it is longer and thinner. This condition is convergent upon that found in the
Plusinae.

33 (0-667) Most outgroup genera have ridged styloconic sensilla apically on the proboscis. In the
plusiines, these sensilla are smooth, as they are in Cucullia, by convergent derivation. The euteliines
are characterized by the further elaboration of three of the ridges into ‘wings’. Close examination of
the sensilla of Stictoptera (Fig. 113) reveals partially-developed ‘wings’, thus strengthening the
hypothesised sister-group relationship between the Stictopterinae and the Euteliinae.

34 (1-000) State 1 for this character unites Magusa, the Stictopterinae and the Euteliinae. State 2 is
found only in Brachionycha and Diloba and is functionally correlated with character states 31:0 and
32:2; the lack of apical styloconic sensilla allowing setae to extend to the apex of the reduced
proboscis.

35 (1-000) Only in the Plusiinae are the strengthening bars of the proboscis semicircular to the apex.
The apical nodules in the euteliines are restricted to the extreme tip of the proboscis and are preceded
by very strong semicircular bars. However, these bars appear to be structurally different to those in
the Plusiinae and are treated as non-homologous.

36 (1-000) Dense, regular, basal microspines on the proboscis are characteristic of the euteliines.

Labial palps
37 (1-000) Smooth ventral edges to the labial palps unite the stictopterines, euteliines and Magusa.
38 (0-500) Although the apomorphic state for this character defines the outgroup clade 51 (less

Oncocnemis + Chalcopasta, in which it is reversed), the semi-erect scales in the four genera
concerned are differently distributed and thus possibly not homologous. In Diloba and Cucullia,
semi-erect scales are present on both the inner and outer surfaces; in Brachionycha, they are
restricted to the inner surface; while in Calophasia, they occur only on the outer surface.

39

40

41

42

43

44

SYSTEMATICS OF THE PLUSIINAE 127

(0-500) A blade-like labial palp segment 3 is characteristic of the Euchalciina but is reversed in
Euchalcia + Adeva.

(0-500) A labial palp in which segment 1 is subequal in length to segment 2 is apomorphic for clade 51
in the outgroups but is reversed in Calophasia.

(0-250) An elongate labial palp segment 3 is interpreted as having been independently gained with-
in the Euchalciina in Polychrysia, Chrysanympha and Lamprotes + Pseudeva. Plusiopalpa also
has a superficially similar elongate segment 3 but in this genus, close examination reveals that
the elongation is dorso-distal to the VRO (Fig. 157) and so not homologous with the general elon-
gation found in the other four genera, where the VRO remains subapical in position (e.g. Fig.
156).

(0-250) An acutely pointed labial palp segment 3 has been independently acquired in Omorphina,
Oncocnemis, Lamprotes + Pseudeva and Panchrysia + Chrysanympha.

(0-500) There are two possible transformation sequences for this character on the cladogram. State 1
can either be independently gained in the Stictopterinae and Paectes, or unite the Stictopterinae and
Euteliinae, with subsequent reversal in Anuga. This latter optimization is favoured by the observa-
tion that, while a third stictopterine, Nigramma, possesses state 1, a fourth, Odontodes, lacks it.
Thus, in this slightly larger sample of genera, state 1 would have to be reversed in Odontodes (as this
genus would form the sister-group of Lophoptera rather than be the basal stictopterine lineage;
Holloway, 1985), and so it seems more likely that reversal has also occurred in Anuga.

(0-500) A labial palp segment 3 that is subspherical appears to have been independently derived in
Brachionycha and in Chalcopasta + Oncocnemis.

Antennae

45
46
47

48

(0-500) Very long ventral hairs on the antennae have been independently acquired in Stictoptera and
Paectes.

(0-667) Asymmetrically bipectinate antennae are characteristic of the Euteliinae, whereas symmet-
rically bipectinate antennae have independently arisen in Brachionycha + Diloba and Lophoplusia.
(1-000) The three states of this character are mutually exclusive to the three groups that have
pectinate antennae: Lophoplusia (state 1), the Euteliinae (state 0), and Brachionycha + Diloba
(state 2). The remaining genera were coded * and, as a result, the optimization procedure allowed
every node in the cladogram between the three groups mentioned to take any value, 0, 1 or 2. Only a
more detailed study of antennal morphology will determine whether any (or none) of these states is
comparable to the condition found in those genera with simple antennae, and thus which, if any, is
plesiomorphic.

(1-000) Pectinations that arise basally on each segment may be synapomorphic for Brachionycha
+ Diloba. Alternatively, pectinations that arise distally could be apomorphic for a large group
of taxa, including the euteliines and plusiines, but are only expressed in a small minority of the
included species. Only the discovery of pectinations in a taxon that normally has simple antennae
might provide the required information to allow a decision to be made as to which state is
plesiomorphic.

External thoracic structure

49
50

Dil
g52

53
54
55
56

(0-500) A large thoracic crest is apomorphic for the Plusiinae but also occurs in Chalcopasta.

(0-500) Vertically-held tegulae appear to have been convergently acquired by Polychrysia and

Pseudeva. This was a difficult character to score, and partially-erect tegulae occur in other euchal-

ciine genera (Lamprotes, Plusidia). More detailed examination may reveal that character 50 defines

a much larger group.

(1-000) A truncate dorsal tegular arm is apomorphic for the Abrostolini.

(0-667) State 2 of this character displays a similar distribution to character 43, except in this instance,

the apomorphic state can either unite the Stictopterinae and Euteliinae then reverse in Stictoptera,

or be independently derived in the Euteliinae and Lophoptera. State 1 is autapomorphic for Omor-
hina.

(0-500) The colourless, microspinose extension to the ventral tegular arm is convergently derived in

Stictoptera and Anuga.

(1-000) Microspines along the inner edge of the ventral tegular arm are synapomorphic for

Calophasia and Cucullia.

(1-000) A ventral tegular arm that is produced on a lateral extension of the main tegular body is

synapomorphic for Brachionycha and Diloba.

(1-000) A broad marginal flange on the dorsal tegular arm may prove synapomorphic for the

128

a7

Legs
58

59

60

IAN J. KITCHING

Plusiinae, Euteliinae, Stictopterinae and Magusa. However, only Brachionycha and Diloba truly
lack a flange; the members of clade 52 possess a very narrow flange, less than one scale-base wide, and
thus further sampling may result in reinterpretation of this character.

(1-000) The attachment point of the patagium is drawn out only in the Stictopterinae + Euteliinae.

(0-111) This character posed coding problems for those genera in which the foretibia was striped or
mottled (Chrysanympha, Adeva, Cucullia, Diloba, Magusa, Stictoptera), or entirely white (Eosphor-
opteryx). These taxa were coded *. This resulted in several alternative sequences of state change in
the Euchalciina. Given that the foretibial white spot is frequently lost elsewhere in the subfamily
(Zonoplusia, Allagrapha, Diachrysia, Plusia), the best optimization assigns state 1 to Adeva and
independent losses to Plusidia and Euchalcia. Chrysanympha and Eosphoropteryx were optimized to
bear state 1. In the outgroups, the white spot was convergently acquired by the euteliines and
Chalcopasta. Cucullia, Diloba, Magusa and Stictoptera were treated as having state 0.

(0-083) The colour of the dorsal surface of the hind tibiae and tarsi is the most homoplasious
character in the data, with six gains and six losses, yet with only one ambiguous assignment of state to
an internal node. The uniform white colour may have been independently gained by Omorphina and
the Argyrogrammatini + Plusiini, or gained by the Plusiinae as a whole and lost in the Abrostolini.
Convergent gains occur in Chalcopasta, Cucullia, Calophasia (following loss in clade 43) and
Autoplusia (following loss in clade 34). The remaining reversals occur in Rachiplusia, Pseudoplusia
and clade 9. Many genera proved difficult to code in that they had a generally unicolorous pattern in
which the white apices to the segments were still faintly visible: Trichoplusia, Zonoplusia, Pseudo-
plusia, Argyrogramma, Agrapha, Adeva and Syngrapha. They were originally coded 0. Had these
taxa been coded 1, the homoplasy would actually have been greater by one step. Character 59 is thus
best dismissed as containing any reliable information.

(0-500) The basitarsal comb on the male hindleg appears at first analysis to be a very labile character
that is interpreted as having been multiply acquired and transformed within the Argyrogrammatini.
State 1 is present in Dactyloplusia, Trichoplusia + Thysanoplusia and Ctenoplusia + Acanthoplusia,
as an independent development in each. State 2 occurs as a transformation of state 1 in Ctenoplusia
and as an independent derivation from state 0 in Plusiotricha. This is not improbable because the
comb in Plusiotricha consists of a single row of about 9 setae, while that in Ctenoplusia comprises only
two setae (but see below). State 3 in Argyrogramma + Stigmoplusia also arises de novo. However,
while two other species of Stigmoplusia (S. allocota Dufay and S. acalypta Dufay) also possess state 3,
two other Argyrogramma (A. basigera and A. signata) lack combs entirely. Additionally, while all
species of Agrapha lack a differentiated comb, in one species (an unnamed species in the BMNH
collection; sp. A, see Agrapha, Clade analysis), the inner SV row of spines is basally longer and
thinner.

For the remaining genera, a more thorough survey was performed. All Acanthoplusia species
examined (N = 4) agreed with A. tarassota. Most Ctenoplusia species (N = 30) had a short row of 2—7
setae, agreeing with C. limbirena in setal form, if not exactly in number. C. aenofusa (Hampson) and
C. phocea (Hampson) exhibited state 3 and this suggests that these two species may not belong in
Ctenoplusia but may be more closely related to Argyrogramma and Stigmoplusia. C. epargyra
(Dufay) and C. leucostigma (Dufay) lack both a tarsal comb and true blade-like setae on the valve
(see character 133), and, likewise, perhaps should be excluded. The former owes its place in
Ctenoplusia (Dufay, 1970a) to a superficial similarity to Ctenoplusia mapongua (Holland) and
Trichoplusia sestertia (Fig. 44) (then in ‘Plusia’ s.1.), while the latter resembles C. phocea in habitus.
Finally, C. albostriata (Bremer & Grey) and C. subchalybaea (Walker) (treated as a synonym of the
previous species by Holloway, 1985), together with Trichoplusia oxygramma and T. vittata, also lack
a tarsal comb. The first three taxa, however, do bear modified setae along the ventral edge of the
valve. Nevertheless, they may not belong in Crenoplusia. If the above anomalous species of
Ctenoplusia are excluded, then the genus becomes extremely well circumscribed by the form of the
tarsal comb. The modified setae may then prove to be synapomorphic for the genera Ctenoplusia,
Acanthoplusia, | Eutheiaplusia| and Plusiotricha (see character 133).

Within Trichoplusia (N = 29, excluding T. oxygramma and T. vittata) and Thysanoplusia (N = 9),
eight species of the former (T. rectilinea (Wallengren), T. sestertia, T. exquisita, T. arachnoides, T.
callista Dufay, T. photeina Dufay, T. dolera Dufay and T. gromieri Dufay) and three species of the
latter (Thy. orichalcea, Thy. florina (Guenée) and Thy. viettei (Dufay)) lack tarsal combs. However,
the vagueness of the limits of these genera renders a critical evaluation impossible until a species level
cladistic analysis has been undertaken.

61

62

63

64

65
66

67
68

69

SYSTEMATICS OF THE PLUSIINAE 129

Overall, it should not be considered unlikely that tarsal combs will be shown to have evolved two or
more times within the Argyrogrammatini, with multiple, independent subsequent losses. Neverthe-
less, state 3 may prove to be synapomorphic for Stigmoplusia, (?part of) Argyrogramma and several
possibly related species (?Ctenoplusia aenofusa, C. phocea), while state 2 may be apomorphic for
Ctenoplusia, or a large subset thereof.

(0-600) State 1 for this character has been acquired independently three times, in the Stictopterinae,
Plusiotricha and clade 17. The first of these is almost certainly convergent upon the plusiine condition
as a result of logical dependence on the following character. Further suppression of the male
foretibial spines (state 2) is synapomorphic for Argyrogramma and Stigmoplusia, while the most
extreme modification is autapomorphic for Argyrogramma.

(1-000) Reduced hind basitarsal spines are apomorphic for the Stictopterinae + Euteliinae, with
total loss on the basal half of the segment occurring in the stictopterines. Spines are also suppressed
on the other legs in species showing state 2 and this gives rise to a correlated coding for these genera
for character 61.

(1-000) Spining on the fore and midtibiae was treated separately from that on the hindtibiae (see
character 64). Apical spines (state 1) were autapomorphic for Rachiplusia, while extensive spining
(state 2) was autapomorphic for Caloplusia.

(0-429) Spines on the inner surface of the hindtibia are autapomorphic for Mouralia and form a
separate transformation series from that concerning spines on the outer surface. Most instances of
the latter appear to have been independently derived: state 1 in Autoplusia, Cornutiplusia and clade
44; state 2in Anagrapha and Rachiplusia. Only in Caloplusia is state 2 derived from state 1. However,
as mentioned above, occasional spines may be observed in Plusia and Autographa, and so may prove
apomorphic for the more extensive clade 38, with multiple subsequent suppression. Also, in
Autoplusia, one leg of the male A. egena examined showed state 2, while the other leg, together with
both legs of the female, displayed state 1. Thus this character is also plastic in Autoplusia. Overall, the
expression of character 64 is variable and should be treated with extreme caution when assessing
degrees of relatedness.

(0-750) The three forms of SD and D setae on the tarsi have been independently derived; state 1 in
Diachrysia, state 2 in Polychrysia, and state 3, convergently, in Panchrysia and Lophoplusia.
(1-000) Hindlegs are shorter than midlegs only in the Stictopterinae + Euteliinae.

(0-500) Ventral lobes on the claw are convergently absent in Oncocnemis and Diloba.

(0-667) Elongate hindleg inner apical spurs occur in the Stictopterinae and Euteliinae. Two
sequences of character state changes are possible. States 1 and 2 can be independently derived in
Stictoptera and the euteliines respectively (the latter via state 1); or state 1 can unite the two
subfamilies with a transformation to state 2 in the euteliines and a reversal to state 0 in Lophoptera.
Further sampling within these two subfamilies is required in order to decide between these two
competing transformation series.

(0-500) Foretibial claws have been independently derived in Brachionycha and Oncocnemis +
Chalcopasta.

Internal structure of the prothorax

70
TA

(1-000) A posteriorly recurved prothoracic furca characterizes the Euteliinae + Stictopterinae.
(0-500) A ridge on the dorsal edge of the epimeron is convergent between the Euteliinae and
Brachionycha + Diloba.

Male abdomen Al—7

2
73

74

75

(0-500) The narrowly tapered abdomen occurs in Stictoptera and the euteliines. As for character 17,
this distribution can be accounted for by a gain followed by a loss, or by two independent gains.
(1:000) State 3 forms an independent transformation, characterizing the Stictopterinae + Eute-
liinae. There is some ambiguity regarding the fitting of states 1 and 2. The most morphologically
parsimonious sequence treats state 1 as apomorphic for the Plusiinae less the Omorphinini, in which
the ventro-lateral edges of the internal flange on A2 are expanded internally. This condition is
retained in the Abrostolini, but in the higher Plusiinae (clade 3), the lobes of the flange are even more
highly developed, meeting medially but with a ventral notch.

(1-000) Internal longitudinal flanges on St2 are autapomorphic for the Euteliinae. Holloway (1985)
stated that these structures may be correlated with the resting position of the adults, in which the
wings are longitudinally folded and the abdomen is recurved dorsally.

(1-000) Components of the trifine hair pencils are found, in this study, only in Cucullia and Magusa,
where they are interpreted as independently derived. Far more taxa need to be examined before any

130

76

77

78

79

IAN J. KITCHING

meaningful information can be extracted from this frequently highly labile (Birch, 19726) character
complex.

(1-000) Postero-lateral sclerotizations on St7 are apomorphic for clade 6 — the Argyrogrammatini
less Anadevidia. They are only just discernible in Pseudoplusia and Eutheiaplusia, a condition that
may be correlated with the reduced hair pencils and subdorsal tufts of these genera (see below).
(0-429) The presence of dorso-lateral hair tufts on segments AS5-—7 characterizes clade 6, with
independent losses in Zonoplusia and Pseudoplusia. Additional tufts on segments A(2—)4 are
synapomorphic for the members of clade 12, but with subsequent loss in Acanthoplusia + Ctenoplu-
sia. The presence of distinct additional tufts on A4 is autapomorphic for Chrysodeixis. As alluded to
above, it may eventually be possible to subdivide state 2 and extract more information, whence
further resolution within the Argyrogrammatini may be forthcoming.

(1-000) As was originally suspected, the three forms of enclosed hair tufts on A7 are independently
derived in Dactyloplusia (state 1), Ctenoplusia (2) and Argyrogramma (3).

(1-000) This character (distinct edges to the tufts on A5 and A6) has been discussed earlier in
connection with the decision between subcladograms 1 and 2.

Male St8 and ventral hair pencils

80

81

82

83

(0-600) This character is now shown to consist of at least two independently derived forms of hair
pencil. State 3 characterizes the Stictopterinae + Euteliinae + Magusa, but is also found in
Calophasia + Cucullia. The hair pencils coded by states 1 and 2 are restricted to the Plusiinae. The
former unites the members of clade 2 (the Plusiinae less Omorphina), while state 2 has been
independently derived in Dactyloplusia and the argyrogrammatine clade 10. There is some evidence
to suggest that the conditions of St8 found in Zonoplusia, Pseudoplusia and Chrysodeixis may
eventually be recognized as reduced forms of state 2, whence this feature would become apomorphic
for clade 6. Several other taxa also show partial reductions of the hair pencils and St8.

(1-000) The plesiomorphic state for the stictopterine type of hair pencil is interpreted as state 3. This
causes some problems in the optimization of states on the internal nodes of the outgroups because
Brachionycha, Diloba, Chalcopasta and Oncocnemis all lack any type of hair pencil and were coded
*. However, given the ambiguity in the outgroup relationships and their relatively distant affinities,
this is not a serious uncertainty in the current study. Within clade 47, state 0 is probably plesiomor-
phic, with states 2 and 1 being autapomorphic for Magusa and Anuga respectively. Given that
Magusa may not represent the immediate sister-group of the Stictopterinae + Euteliinae (see clade
47, Clade analysis), this sequence of state transformation seems more prudent than the alterna-
tives.

(1:000) Similarly, character 82 does not apply to clades 53 and 55. Thus state 0 is apomorphic for
Calophasia + Cucullia, and state 1 for clade 47. State 2 is autapomorphic for Stictoptera. Anuga,
coded *, is predicted to have state 0.

(0-313) This character, the degree of development of the male sternal hair pencils, proved, as
expected, to be highly homoplasious, with numerous alternative character state optimizations. The
sequence chosen assumed that state 0, as seen in Abrostola, is plesiomorphic for the Plusiinae (less
Omorphina). The large single pocket (state 5) of Mouralia is autapomorphic within the Abrostolini.
The initial transformation 0 — 3 (well developed, separate pockets) characterizes the Argyrogram-
matini + Plusiini. Within the Plusiini, the pockets are reduced to state 0 again in the Euchalciina. This
confirms that state 0 in this subtribe is the result of secondary loss and is not due to plesiomorphic
absence, as in Abrostola, providing that hair pencils have not been independently gained in the
Argyrogrammatini and Autoplusiina + Plusiina. A change to state 2 (partially fused pockets)
characterizes clade 30 (the Autoplusiina less Rachiplusia) but with a reversal to state 3 in Autoplusia
itself. The Plusiina are characterized by the change 3 — 1 (reduction of pockets to slight invagina-
tions), with independent reversals to states 3 and 0 in Plusia and Syngrapha respectively.

The situation is less simple in the Argyrogrammatini. A parallel change to state 2 (as in clade 30)
occurs to unite the members of clade 6. Within clade 8, there are independent transformations to
states 0 and 5 in Pseudoplusia and Chrysodeixis respectively. The form of the hair pencil is then even
more highly modified in the more derived members of the latter genus, as outlined above (Terminol-
ogy). State 4 (fused pockets) is apomorphic for clade 17, with a reversal to state 3 in Stigmoplusia.
There are also parallel changes from state 2 to state 3 in Acanthoplusia and Trichoplusia +
Thysanoplusia.

The sternal hair pencils are obviously highly variable and maybe labile structures. However, the
distribution of states and homoplasy is very far from random and more extensive sampling could
result in this complex proving very useful.

SYSTEMATICS OF THE PLUSIINAE 131

84 (0-250) A sclerotized bar on the hair pencil pockets is apomorphic for clade 10 but with subsequent
independent losses in Plusiopalpa, Acanthoplusia and Thysanoplusia.

85 (1-000) Differentiated hair pencil insertions are synapomorphic for Argyrogramma and Stigmoplu-
sia.

86 (0-333) As discussed above, accessory pockets on the preferred cladogram must be developed in
clade 14 but then independently lost in Plusiopalpa and Ctenoplusia.

87 (0-500) Short, outer scales on the hair pencils also characterize clade 14, but with a reversal in clade
a7.

88 (0-500) Long inner tufts of hair pencil scales are convergently derived in Agrapha and Argyro-
gramma.

89 (1-000) Secondary hair brushes are synapomorphic for the members of the Argyrogrammatini.

90 (1-000) An anterior flange on the lateral arms of St8 is restricted to Argyrogramma and Stigmo-
plusia.

91 (0-500) Lateral arms of St8 that twist dorsally over the hair pencils are apomorphic for the

Argyrogrammatini less Anadevidia, but the change is reversed in Pseudoplusia, which also has much

reduced hair pencils. Further, the development of a deep basal concavity characterizes clade 12 but

with subsequent reversal in Plusiopalpa.

(0-750) This character was also discussed in detail above, with regard to subcladograms 1 and 2.

State 1 is convergently derived in clade 12 and Dactyloplusia; state 3 is apomorphic for clade 14 and

this transforms to state 2 to unite Argyrogramma and Stigmoplusia.

93 (0-600) The form of fusion of the apices of the dorsal arms of St8 is fairly homoplasious. State 2 (in
which the arms fuse into a rounded apex) is apomorphic for clade 10, with further transformations to
state 3 (transversely oriented dorsal arms), uniting Argyrogramma and Stigmoplusia, and state 1
(converging dorsal arms), autapomorphic for Thysanoplusia. However, state 1 occurs outside clade
10 in clade 7, with a reversal to state 0 in Pseudoplusia. This character has probably not been coded
correctly and further sampling of taxa may clarify matters. The form of St8 will almost certainly prove
to be highly important in the future critical resolution of the interrelationships of the taxa within the
Argyrogrammatini.

94 (0-750) There are several alternative optimizations for the internal node states for this character too.
The most probable, on present evidence, interprets state 2 (presence of broad, flat, ‘greasy’-looking
scales ventrally between the dorsal arms of St8) as apomorphic for clade 6. There are subsequent
transformations to state 3 (scales fused into a flat plate) in Argyrogramma + Stigmoplusia, to state 0
(a reversal) in Pseudoplusia, and to state 1 (broad, erect, spatulate scales) in Chrysodeixis (which is
further modified in the more derived species of this genus; see above). Superficially similar scales to
those in Chrysodeixis occur in Anadevidia, but in the latter genus, they were interpreted as the scales
of the secondary hair brush. Thus they are convergent upon the condition found in Chrysodeixis,
which bears normal secondary hair brushes in addition to the apical spatulate scales.

95 (0-500) Long, fine scales ventrally between the lateral arms of St8 characterize clade 30, although
they are subsequently lost in Sclerogenia.

9

Nw

Male genitalia — uncus, tegumen & anal tube

96 (0-500) A ventrally flattened uncus is considered to have arisen independently in Paectes, Calopha-
sia and Brachionycha + Diloba. State 2, a subspherical uncus, is autapomorphic for Anuga.

97 (0-250) A straight uncus has been derived independently four times: Chalcopasta, Chrysodeixis,
Stigmoplusia and clade 44.

98 (0-500) The uncus has become distally dilated dorsally independently in Agrapha and Argyro-
gramma.

99 (1-000) A notch between the tegumen and uncus is characteristic of the Plusiinae less the Omorphi-
nini. A further production of the tegumen apex unites Allagrapha, Macdunnoughia and Puriplusia.

100 (1-000) A concave base to the uncus is an apomorphy of the Abrostolini.

101 (1-000) A tegumen with a distinct angle is synapomorphic for the members of the Euchalciina. The
apomorphic state is weak in Panchrysia and Pseudeva, and present in only one of the two Euchalcia
specimens examined.

102 (1-000) An uncus that is medially flattened and expanded is apomorphic for the Abrostolini.

103 (1-000) The presence of a distinct median suture on the uncus is also apomorphic for the Abrosto-
lini.

104 (0-500) Widely divergent tegumenal arms are convergently derived in Antoculeora and Erythro-

lusia.

105 (0-111) The origin of the anal tube with respect to the LASTs and uncus base is highly homoplasious.

132

106

107

IAN J. KITCHING

This character seems generally to be correlated with the overall dorsal appearance of the tegumen
apex, LASTs and uncus base. There is a general trend, which is difficult to quantify, that as the anal
tube moves anteriorly, the apex of the tegumen becomes elongate and the outline of the tegumen >
LASTs — uncus smoothes out. Until such time as the morphology of this complex region is clarified,
the best optimization for the development of the apomorphic state may be to postulate independent
gain wherever possible. Thus, state 1 may have arisen separately in Chalcopasta, clade 38 (with
reversal in Plusia), Anadevidia, Agrapha, Argyrogramma, Trichoplusia and clade 7. In the last of
these, the anal tube of Chrysodeixis chalcites arises at the uncus base but this condition appears to be
the end product of a reversal series. The ‘plesiomorphic’ species, C. argentifera, has an anterior
origin of the anal tube (state 1), as in Dactyloplusia and Pseudoplusia, while the ‘intermediate’ C.
illuminata shows an intermediate position. The ‘derived’ C. kebea (and C. chalcites), as defined by an
arrowhead-shaped St8 (see Terminology), have reverted totally to the plesiomorphic condition for
character 105.

(0-500) The fusion of the LASTs to the underside of the tegumen has occurred independently in the
Argyrogrammatini + Plusiini and in Chalcopasta.

(0-200) Subdorsal bars on the anal tube have been gained on several occasions: Chrysanympha,
Lophoplusia, the Abrostolini, and clade 47 (with subsequent loss in the euteliines). Within the latter,
an alternative optimization hypothesises independent gain in Magusa and the stictopterines.
However, extreme modification of the LASTs in Paectes and their complete absence in Anuga
suggests that loss of the bars in the euteliines is a more satisfactory hypothesis.

Male genitalia — vinculum & pleurite

108

109

110

iI

112

113

114

115
116

117

(0-125) The presence of a ventral tooth on the vinculum is also a highly homoplasious character
state. It has been acquired by clade 39 (with no reversals). Erythroplusia, Sclerogenia and the
Argyrogrammatini. Within the latter, the tooth has been lost four times, in clades 7 and 13, and in
Acanthoplusia and Argyrogramma. The tooth represents the insertion point of the S2 muscle (Bayer,
1965), which, together with other muscles, retracts the genitalia into A8. Bayer noted the absence of
$2 in Chrysodeixis acuta, a species that also lacks a tooth. Thus, an extensive investigation of genital
musculature, and its correlation with the presence/absence of the vincular tooth, might help clarify
the homoplasy exhibited by character 108. (For an alternative musculature nomenclature, see
Tikhomiroy, 1979.)

(0-333) A vinculum that is S-shaped in lateral view has been independently gained in clades 21
(weakly expressed) and 39 and in Cucullia. Since the S1 muscle inserts dorso-apically on the vinculum
(Bayer, 1965), this shape could also be functionally correlated with the musculature for retracting the
genitalia.

(0-200) A Y-shaped vinculum has been convergently derived in Paectes, Rachiplusia and clade 6
(with subsequent reversal in Dactyloplusia and Trichoplusia). This character is relatively trivial and
probably of low importance.

(0-500) An attenuated spine-like vinculum is apomorphic for Chrysodeixis + Pseudoplusia. Stigmo-
plusia has a produced vinculum but with a rounded apex, a condition shown here to be non-
homologous.

(0-750) Dorso-lateral emarginations (state 1) of the vinculum arms have been independently
acquired by Thysanoplusia and clade 14. Within the latter, the emarginations are expanded into
sharp points (state 2) in Stigmoplusia. The blunt hooks (state 3) of Plusiotricha represent an
independent, non-homologous derivation from state 0.

(0-333) A broad, flat, cone-like vinculum is regarded as apomorphic for the members of clade 31 but
is reversed in the included clade 35. It is also independently derived in Anagrapha (but see Clade
analysis).

(0-500) A flattened, entire, conical vinculum is convergently derived in Rachiplusia and the
stictopterines + euteliines. Originally scored *, Stictoptera was optimized to possess state 1. This
suggests that further work in the Stictopterinae should show the condition in this genus to be
derivable from state 1.

(1-000) Distally convergent vinculum arms are apomorphic for the Euteliinae.

(0-333) A distinct, sharp, basal notch on the pleurite is apomorphic for the Autoplusiina + Plusiina,
although it is convergently present in Anadevidia and reversed in Antoculeora. The latter condition
was commented upon above, when discussing the position of Rachiplusia and the branching
sequence of the subtribes of the Plusiini, and is interpreted as being correlated with the highly
modified male genitalia found in Antoculeora (see also characters 117 and 118).

(0-500) A broad, flat lateral flange on the pleurite is considered to be an apomorphy of the Plusiini.

SYSTEMATICS OF THE PLUSIINAE 133

Like character 116, the apomorphic state is lost in Antoculeora. According to Bayer (1965), this
flange represents the insertion point for the genitalia retraction muscle $3.

118 (0-667) State 1, in which the pleurite lies along the antero-dorsal edge of the tegumen, is interpreted
as plesiomorphic. State 2, where only the apex of the pleurite is in contact with the tegumen, is
autapomorphic for Antoculeora, while Magusa and Calophasia appear to have independently
acquired state 0, in which the apex of the pleurite articulates on the inner edge of the tegumen in a
pocket. However, * codings for Diloba, the stictopterines and the euteliines render this conclusion
open to question.

119 (0-500) Partial fusion of the pleurite and VAVB has convergently arisen in Plusiotricha and
Ctenoplusia + Acanthoplusia (but see characters 60 and 133, and Clade analysis).

120 (1-000) A very narrow pleurite closely appressed to the antero-dorsal edge of the tegumen is
synapomorphic for Chrysodeixis and Pseudoplusia.

121 (0-500) The three states of this character are treated as independent derivations, all of which occur
twice. A very short or missing VAVB (1) characterizes the Plusiini and Stictopterinae. A VAVB
modified into a blunt lobe (2) arose convergently in Abrostola and Argyrogramma. In Calophasia
and Oncocnemis, the VAVB is missing and the tegumen base is produced below the pleurite to
approach the vinculum apex (3).

122 (1-000) Socii are apomorphic for the euteliines.

123 (0-500) Complete fusion of tegumen and vinculum arms has, as suspected a priori, occurred in
independently in Diloba and the Stictopterinae + Euteliinae.

Juxta & anellus

124 (1-000) The apex of the juxta is fused to the aedeagus ventrally only in the Abrostolini.

125 (1-000) Apical thickening of the juxta is also apomorphic for the Abrostolini.

126 (1-000) Fusion of the juxta, by a bar, to the inner edge of the sacculus near the base of the clavus
occurs only in the Abrostolini.

127 (0-429) A parallel-sided notch has developed apically on the juxta three times in the Plusiinae; in
Omorphina, Chrysodeixis and the Plusiini. Within the latter, it has been lost in Allagrapha and
independently modified into a subapical process twice, in Anagrapha and clade 44. The further
elaboration of this process into a sharp spine is synapomorphic for Syngrapha and Caloplusia.
Although the character is homoplasious overall, state 3 may allow some resolution within Syngrapha
(s.1.). The sharp spine is present only in a subset of those species with yellow hindwings (see character
7) allied to S. devergens and Caloplusia hochenwarthi: S. ignea, S. microgamma, and S. alticola. The
spine is absent in S. montana and represented by a ‘strong point, but not a distinct, sclerotized spine’
in S. parilis (Eichlin & Cunningham, 1978). However, similar structures occur elsewhere; S.
angulidens (Smith) has a sharp apical point, S$. sackenii shows a similar structure to S. parilis, and
many other species exhibit a weak, blunt projection. Nevertheless, used with caution, the form of the
juxta apex should prove useful in elucidating relationships within Syngrapha and maybe the Plusiina
as a whole.

128 (0-400) A scobinate juxta apex is apomorphic for clade 7, with distinct thorns autapomorphic for
Pseudoplusia. State 1 has also been independently derived in Thysanoplusia and, as initially
suspected, Plusiopalpa and Cucullia.

129 (1-000) A weakly sclerotized, crescentic juxta is autapomorphic for the Euteliinae.

130 (0-333) For this character, the outgroups were all coded *. The plesiomorphic state in the Plusiinae
was interpreted to be state 1, in which the juxta is fused to the saccular flanges. The juxta has become
free three times independently, when saccular flanges are also present, in Agrapha, Pseudoplusia +
Chrysodeixis and Argyrogramma + Stigmoplusia. Interestingly, all those plusiine taxa that lack
saccular flanges (Abrostolini, Trichoplusia, Autoplusia) were optimized to display state 1 — juxta
fused to flanges! This result suggests that a recoding of this character, incorporating the next (q.v.),
will be necessary if any sense is to be made of the feature in future studies.

Male genitalia — valve

131 (0-250) The presence of saccular flanges characterizes the Plusiinae, with independent losses in the
Abrostolini, Trichoplusia and Autoplusia. The implication for the previous character is that, had
saccular flanges not been lost in these taxa, then they would have been fused to the juxta.

132 (1-000) The presence of a rounded lobe between the valve bases, formed from the fused saccular
flanges, is synapomorphic for Panchrysia and Chrysanympha.

133 (0-500) Blade-like setae are interpreted here as convergently derived in Ctenoplusia + Acanthoplu-
sia and Plusiotricha. As noted above (character 60), there appear to be problems regarding the

134

134
135
136

1257)

138

139

140
141

142

143

IAN J. KITCHING

homologies of this and similar setal modifications. In the present context, the setae coded by state 1
are strongly sclerotized, regularly arranged in a single row along the ventral edge of the valve and
shaped like short, broad, curved blades. In Ctenoplusia albostriata and Trichoplusia oxygramma
(McDunnough, 1944), the ventral valval setae are weakly sclerotized and narrow, and thus may not
be homologous with the setae found in typical ctenoplusias. Further work is necessary to decide
whether this group of species truly belongs in Ctenoplusia or represents, say, the sister-group of
Ctenoplusia (+ other genera?). Similarly, C. epargyra and C. leucostigma only have spiniform hairs
on the ventral edge of the valve, which, in conjunction with their lack of a tarsal comb, argues for
their exclusion from the genus. The setae in C. phocea are also weak and narrow. Additionally, they
are in multiple rows basally, unlike other Crenoplusia species. C. phocea also has a tarsal comb
consisting of multiple rows of setae (60:3), an elongate vinculum (111:1) but with a rounded apex,
and moderately strong dorso-lateral emarginations on the vinculum arms (112:1), strongly suggest-
ing a relationship with Stigmoplusia. Finally, C. aenofusa is enigmatic. It has highly modified and
autapomorphic setae along the ventral edge of the valve. They are strongly sclerotized and in a single
row. However, the distal setae are narrow and strongly angled, while the basal setae are apically
spatulate. Also, the setae on the basal half of the sacculus are borne on small marginal projections,
unlike most ctenoplusias. As stated above, the tarsal comb is of the form found in Argyrogramma and
Stigmoplusia, and C. aenofusa, too, may have its true affinities with these genera.

All Acanthoplusia species possess blade-like setae. Dufay (1970b) and Holloway (1985) state that
these are missing in A. vermiculata Dufay, but recessed scale bases along the ventral edge of the
valve, typical of Ctenoplusia and Acanthoplusia (and also Plusiotricha), are visible in the holotype. It
must be assumed that the blade-like setae, which are often caducous in many species, have been
accidentally removed before or when the specimen was prepared and mounted.

Two other genera in clade 14 have modified valval setae. In Stigmoplusia, although the type
species lacks such setae, three other species (S. allocota, S. acalypta, S. megista Dufay) all possess
single, short, conical, somewhat flattened setae apically on the valve digitations. In Agrapha, the
ventral edge of the valve is dorso-ventrally flattened and covered with a dense mat of short, upright,
curved, flat scales.

There is obviously a great deal of useful information in these valval setae within clade 12 but an
exhaustive species-level analysis will be necessary to extract the maximum information.

(1-000) A spatulate lobe arising from the sacculus is autapomorphic for the Abrostolini.

(0-500) A uniseriate corona has been convergently derived in Magusa and clade 52.

(0-500) The presence of basally-directed setose scales on the valve apex is synapomorphic for the
Argyrogrammatini. Similar scales in Mouralia are interpreted as convergently derived. In Anadevi-
dia peponis, the valval setae are extremely persistent and very numerous. However, basally-directed
scales are present in A. hebetata (Butler) (Chou & Lu, 1978: fig. 2), in the position occupied by the
caducous setae of the other argyrogrammatine genera, whereas no such setae occur anywhere in the
Plusiini.

(0-500) Setae arising from ventral marginal prominences on the valve are interpreted as apomorphic
for clade 32, but with an independent derivation in Omorphina.

(0-500) A spatulate apex (cucullus) to the valve is treated as synapomorphic for the members of
clade 12 but with reversal in clade 17. The weak nature of the subapical constriction in Eutheiaplusia,
together with weak narrowing in Trichoplusia, Thysanoplusia, Dactyloplusia and Pseudoplusia,
suggests further sampling may reveal a very gradual transformation series leading from an uncon-
stricted valve to the extreme condition seen in Crenoplusia.

(0-500) The clavus base is obscured by an overfold of the sacculus in Rachiplusia and clade 32
convergently.

(1-000) A broad, blunt point on the ventral edge of the valve unites Plusiotricha and Eutheiaplusia.
(0-250) The saccular spine is convergently derived in Diloba and clade 30, but within the latter, it has
been lost in Autoplusia and Loboplusia. The absence in Autoplusia egena is almost certainly a true
loss (rather than plesiomorphic absence) since the spine is present in both A. olivacea and
‘Syngrapha’ gammoides.

(0-333) A saccular sheet, in some form, is characteristic within the Plusiinae of clade 31, although it
has subsequently been lost in Loboplusia. Further modification to form a hook unites Sclerogenia,
Autoplusia and Antoculeora. Convergent occurrences of an apical saccular process appear in
Lophoptera, Calophasia and Magusa.

(0-500) The form of the clavus (characters 143-145) is highly homoplasious. Its presence on a
distinct peninsula (state 3) is probably plesiomorphic for the Plusiinae. The peninsula is convergently
lost in the Abrostolini and Anagrapha, so that the dorsal and ventral margins of the clavus are more

SYSTEMATICS OF THE PLUSIINAE 185

or less level (state 2). However, this state arose de novo from state 0 in Cucullia. The clavus is further
modified into a broad lobe in Abrostola, but this is convergently derived from an unmodified saccular
margin in Chalcopasta + Oncocnemis.

144 (0-133) This character, the degree of development of the clavus in the Plusiinae, was that used by
McDunnough (1944) to subdivide the subfamily into three genus groups. The more extensive
sampling of genera of the present study has shown this view to be far too simplistic. A fully
developed, tubular clavus (state 0) is interpreted as plesiomorphic, because it is the form that occurs
in the only outgroup with a distinct clavus (Cucullia). Multiple, independent reduction seems to be
the rule, but some reversals (i.e. clavi becoming longer and better developed) do occur. Reduction to
a short tubular clavus has occurred in the Argyrogrammatini, clades 24 and 43, Anagrapha,
Loboplusia, Stigmoplusia (after a reversal in clade 15, see below) and Omorphina. Further
reductions to a setose bump have taken place in clade 26, Lophoplusia, Plusiotricha and Agrapha.
Reversal from state 2 to 1 occurs only in Lamprotes, or alternatively (and this, surprisingly, is the only
internal node state ambiguity), state 2 could have arisen independently in Plusidia and Pseudeva.
However, as Lamprotes exhibited a clavus form somewhat intermediate between states | and 2 (see
Characters), this is not too problematical. Reversals from state 1 to 0 occur only in the Argyrogram-
matini but thrice, in Pseudoplusia + Chrysodeixis, Trichoplusia + Thysanoplusia and in clade 15. As
seems to be true of all the widely quoted ‘classic’ key characters (see also, for example, characters 6,
15 and 64), the form of the clavus is highly unreliable as an indicator of cladistic relationship.

145 (0-333) An elongate, flattened peninsula has been convergently derived in Erythroplusia, Macdun-
noughia + Puriplusia and Antoculeora. Interestingly, however, all four of these genera are members
of the Autoplusiina.

146 (1-000) A blade-like valve is apomorphic for the Plusiina.

147. (0-500) More or less oval valves are apomorphic for clade 21 of the Euchalciina but also occur
convergently in Omorphina.

148 (0-200) The presence or absence of the harpes is also very homoplasious, with presence being
interpreted as plesiomorphic. Five independent losses appear to have occurred, in the euteliines,
Stictoptera, Rachiplusia, Anadevidia and clade 34. An alternative optimization in clade 48 hypothe-
sises loss as uniting the Stictopterinae and Euteliinae, with a subsequent regain in Lophoptera.

149 (0-250) It was suggested a priori that the various forms of inwardly-pointing hooked harpes would
prove to be convergently derived and therefore probably non-homologous. This result was obtained,
with independent derivations occurring in clade 44, Trichoplusia, Dactyloplusia and Agrapha.

150 (1-000) Inwardly-curving hooked harpes are apomorphic for Macdunnoughia and Puriplusia, with

: further development of a basal, shorter, upcurving hook autapomorphic for the former genus.
Interestingly, Macdunnoughia crassisigna (Warren) shows an intermediate condition in which the
second hook is represented by a small sharp point.

151 (1-000) Deeply cleft valves are autapomorphic for the Stictopterinae.

152 (1-000) Trapezoid valves are autapomorphic for the Euteliinae.

153 (1-000) Valves that, when closed, lie ventrally or subventrally are synapomorphic for the Euteliinae
and Stictopterinae.

154 (0-333) Transtillae that are fused and produced posteriorly to a broad, blunt point are convergently
derived in Rachiplusia and clade 22 (with reversal in the included clade 26).

155 (0-500) Bifurcate transtillae, discussed earlier in connection with the position of Loboplusia, are
hypothesised to have arisen independently in Erythroplusia and Sclerogenia.

Male genitalia — aedeagus

156 (0-500) A thin, small aedeagus is apomorphic for clade 47 but reversed in Paectes.

157 (0-667) A bulbous coecum (state 2) is synapomorphic for the Plusiini and Argyrogrammatini. In the
latter tribe, the coecum can become very enlarged (e.g. ‘Plusia’ megaloba Hampson [Holloway,
1985: 300]; Ctenoplusia aenofusa [Chou & Lu, 1978: fig. 4]), possibly in order to accommodate an
extremely long vesica when retracted (Holloway, 1985). A dorso-basal origin of the cuticular simplex
appears to have been convergently acquired by Magusa and the euteliines.

158 (0-400) Restriction of the aedeagus sclerotization to a ventral band is synapomorphic for the
Argyrogrammatini and Plusiini, although there has been a reversal to a uniformly sclerotized
aedeagus in Autoplusia + Antoculeora. Further reduction of the sclerotization to a narrow,
well-defined ventral strip is apomorphic for the Argyrogrammatini but has been independently
acquired by the autoplusiine genera Macdunnoughia and Erythroplusia.

159 (0-500) A vesica that is entirely covered with small spines unites Plusiotricha and Eutheiaplusia, but
also occurs in Thysanoplusia.

136
160

161

162

163
164
165
166
167
168
169
170

171
WZ

is

174
175

IAN J. KITCHING

(1-000) A strongly serpentine inverted vesica is also synapomorphic for Eutheiaplusia and Plusio-
tricha.

(1-000) The presence of a small oval sclerotized plate at the base of the ductus ejaculatoris is
synapomorphic for the Stictopterinae and Euteliinae.

(0-111) The distribution of states of this character was discussed in part above in connection with the
positions of Agrapha and Plusiopalpa. The presence/absence of a subapical cornutus (SAC) is highly
homoplasious, with several ambiguous assignments of character states to the internal nodes.
Initially, the presence of a SAC is synapomorphic for the Argyrogrammatini + Plusiini. However,
there have been numerous independent losses, in clades 7, 30 and 44, and in Lamprotes and
Eosphoropteryx. Loss can also be interpreted as having occurred in clade 12, with subsequent regains
in Agrapha and Argyrogramma + Stigmoplusia, but several alternative sequences of change are
possible, including independent losses in Plusiotricha + Eutheiaplusia, Ctenoplusia + Acanthoplusia
and Plusiopalpa. Given that Eutheiaplusia may have a SAC (see Characters) and many ctenoplusias
definitely do (see 648-step cladograms: Agrapha and Plusiopalpa), multiple independent losses may
represent the most likely events.

(1-000) An apically swollen SAC is synapomorphic for Panchrysia and Chrysanympha.

(1-000) A SAC arising from a circular concave plate is synapomorphic for Euchalcia and Adeva.
(0-500) A SAC originating on a lateral diverticulum has arisen convergently in Polychrysia and
Diachrysia.

(1-000) This character (grooved/‘winged’ SAC) was discussed above in detail with regard to the
branching sequence of Plusia, Cornutiplusia and Autographa.

(1-000) The presence of an invaginated sac in the vesica is autapomorphic for clade 6.

(1-000) A vesica in which the base is internally granular is autapomorphic for clade 24.

(1-000) Only the Abrostolini possess a ventral, apical, blunt point to the vesica sclerotization. This is
autapomorphically modified into a hook in Abrostola.

(1-000) A large membranous RHS diverticulum on the vesica, bearing cornuti is synapomorphic for
Abrostola and Mouralia.

(1:000) A dorsal patch of curved cornuti is synapomorphic for the members of clade 26.

(0-500) The presence of two, short, straight cornuti on the vesica is apomorphic for clade 7, although
there has been reversal in Chrysodeixis. In this genus, the position of the short cornuti is occupied by
a series of large cornuti. Further studies are required to determine whether these two forms of cornuti
are homologous.

(1-000) A differentially sclerotized band of RHS sclerotization on the vesica is autapomorphic for
the Argyrogrammatini.

(1-000) A dorsal convex rounded plate on the vesica is synapomorphic for the members of clade 7.
(0-333) Auricular cornuti have been convergently derived in Eosphoropteryx, Loboplusia and
Macdunnoughia.

Female abdomen Al—7

176

177

178

(0-333) A thickened pleuron on A7 is synapomorphic for the members of clade 31 (with an uncertain
loss in Loboplusia — see Characters), but with a convergent derivation in Brachionycha.

(0-500) Concave lateral edges to T7 are also apomorphic for clade 31 (but with a subsequent reversal
in Puriplusia), and thereby confirm the possible correlation with the previous character.

(0-667) A pleural pouch is interpreted as being synapomorphic for the Plusiinae, Stictopterinae and
Euteliinae, but with secondary loss in Magusa. The latter interpretation is supported by the presence
in Magusa of a slight pouch, although this is not enlarged as in the three subfamilies. However, this
conclusion must, for the present, be treated with caution for an alternative optimization of states
interprets the pleural pouch as an independent derivation in the Plusiinae and in the Stictopterinae +
Euteliinae. A small, concealed pleural pouch is autapomorphic for the Euchalciina.

Anal papillae and AS—A9 + 10 intersegmental membrane

(1-000) Flared anal papillae are synapomorphic for the Euteliinae and Stictopterinae.

(0-500) The Stictopterinae + Euteliinae and Calophasia have convergently acquired densely setose
inner surfaces to the anal papillae.

(1-000) Frayed edges to the anal papillae are autapomorphic for the Stictopterinae.

(1-000) A differentiated dorso-ventral row of setae on the anal papillae is an autapomorphy of the
Plusiinae.

(0-500) Short anal papillae characterize clade 22, but with reversal in Euchalcia. Several species of
Euchalcia (e.g. E. italica (Staudinger), E. dorsiflava (Standfuss), E. augusta (Staudinger)) have even

184

185
186

A,
187
188
189
190

191

192
193
194

195
196

197
198

199
200
201
202

SYSTEMATICS OF THE PLUSIINAE 137

more modified anal papillae, in which the apices are ventrally produced into acute points (Dufay,
1968). Although the biology of these taxa is unknown (Dufay, 1968), similar modifications in the
heliothine genus Schinia Hubner are associated with oviposition deep in flowers and unopened buds.
Thus state 0 in Euchalcia variabilis may merely represent an early stage in a reversal transformation
series from short anal papillae to the highly-derived, blade-like forms.

(0-500) The presence of an eversible dorsal pouch between the anal papillae and A8 can be
interpreted either as autapomorphic for the Plusiinae, with reversal in the Abrostolini, or as
independently derived in the Omorphinini and the Argyrogrammatini + Plusiini.

(1-000) A sub-basal dorsal flange on the posterior apophysis is autapomorphic for clade 6.

(1-000) Thickened membrane between the LPV and the anal papillae is synapomorphic for Plusidia,
Lamprotes and Pseudeva.

ostium bursae & antrum
(1-000) A subdorsal, longitudinal, desclerotized band on T8 is autapomorphic for the Abrostolini.
(1-000) A ventral pocket on the anterior edge of T8, dorsal to anterior apophysis, is also autapomor-
phic for the Abrostolini.
(0-250) T8 has become irregularly cut back antero-subventrally below the origin of the anterior
apophysis independently in Anagrapha, clade 44 and clade 30 (but with reversal in Antoculeora).
(0-500) Anterior apophyses have been independently lost in Lophoptera and the euteliines, or lost in
the Stictopterinae + Euteliinae and regained in Stictoptera.
(0-125) ‘Protected’ ostia bursae, as suspected (see Characters), have been acquired independently
no less than eight times, in Brachionycha, Mouralia, Diachrysia, Rachiplusia, Antoculeora, Lam-
protes, clade 32 and clade 44. When the true homologies of these structures have been determined,
more information may be forthcoming but at present this character is useless.
(1-000) Spinose pads lateral to the ostium bursae unite Macdunnoughia and Puriplusia, with the
further development into recurved spinose grooves autapomorphic for the latter genus.
(1-000) Short, dense, upright scales near the mouth of the antrum are synapomorphic for the
members of clade 30.
(0-667) As initially suspected, setae along the inner ventral edge of the antrum (state 2) are
convergently derived in Diloba and clade 26. Lamprotes is autapomorphic in that the setae are
restricted to a subventral and lateral position (state 1).
(0-500) A ventro-anterior extension of T8 shows the same distribution of states as character 190 and
thus has the same two alternative sequences of state change.
(0-167) An open, shallow antrum has been independently acquired on a total of six occasions, in
Diloba, Cucullia, Chalcopasta, Agrapha, Pseudoplusia + Chrysodeixis and clade 21. A more
thorough analysis of the homologies of the states of this character may reduce the homoplasy and
allow more reliable information to be extracted from it.
(0-500) An entirely membranous antrum has been convergently derived in Stictoptera and Cteno-
plusia + Acanthoplusia.
(1-000) State 1 of this character, in which the ventral edges of T8 are extended towards the midline,
unites the members of clade 35. Further extension (state 2) is synapomorphic for the included genera
Autoplusia and Antoculeora. State 3 (T8, antrum and LPV entirely fused ventrally) arises de novo
from state 0 in Rachiplusia.
(1-000) The anterior apophysis arises level with the ventral edge of T8 only in Autoplusia and
Antoculeora.
(1-000) A heavily sclerotized, smooth, open antrum is also synapomorphic for Autoplusia and
Antoculeora.
(0-200) Thick, white membrane lateral to the ostium bursae has arisen five times independently in
Plusiopalpa, Allagrapha, Lamprotes, Euchalcia and clade 44.
(0-500) This character (T8 represented by a thin band dorsally) has the same distribution and choice
of state change sequences as 190 and 195.

Ductus bursae and corpus bursae

203

204

(0-667) A long, narrow ductus bursae is interpreted as apomorphic for clade 9 but with a reversal in
the included clade 14. Trichoplusia is autapomorphic for an extremely long ductus bursae (state 2).
Given that the females of Plusiotricha are currently unknown, and both Trichoplusia and Thysano-
plusia are heterogeneous for states 2 and 1 respectively, very little confidence can be placed in this
character at present.

(0-500) A corpus bursae that is produced at right angles to the long axis of the ductus bursae shows

138 IAN J. KITCHING

the same distribution and heterogeneity of states as the last, and is therefore subject to the same
reservations.

205 (0-200) Thickened white membrane on the corpus bursae at the entry point of the ductus bursae has
been independently acquired by Cucullia, Trichoplusia, [Plusiotricha +] Eutheiaplusia and clade 24
(but with subsequent loss in Lamprotes + Pseudeva).

206 (0-500) An expanded portion of the ductus bursae with dorsally recurved edges has been con-
vergently derived in Rachiplusia and Macdunnoughia + Puriplusia.

207 (0-500) A short, lateral pouch on the ductus bursae occurs independently in Agrapha and Chry-
sodeixis.

208 (1-000) A small corpus bursae with highly elastic walls is synapomorphic for Diloba and
Brachionycha.

209 (0-182) The position of the origin of the ductus seminalis on the corpus bursae is highly homopla-
sious and highlights the need alluded to above for a critical study of the homologies of the structures
variously referred to as the fundus, apex and appendix bursae. The plesiomorphic condition appears
to be an apical origin. Independent shifts to a fundular origin have occurred in Erythroplusia,
Puriplusia, Loboplusia, Panchrysia, Oncocnemis, Cucullia and the Argyrogrammatini. Within the
last, reversal has occurred in Pseudoplusia and clade 10, with re-reversal in /Plusiotricha +]
Eutheiaplusia. However, at least two of these homoplasies (Puriplusia and Loboplusia) may be due
to errors in deciding which end of the corpus bursae represents the fundus and which the apex (see
also Ichinosé, 1962b). For the present, character 209 is best dismissed as unreliable.

210 (0-667) Regular ridges on the corpus bursae have been convergently derived in clade 47 and Cucullia
+ Calophasia. Further modification to form a more or less circular pattern unites the Stictopterinae
and Euteliinae.

211 (0-667) Aconcave, circular signum unites the Stictopterinae and Euteliinae but is lost in Anuga and
further modified into a ridge in Lophoptera.

212 (1-000) A distinct differentiation of sclerotization of the corpus bursae granulations is synapomor-
phic for Sclerogenia, Autoplusia and Antoculeora.

213 (0-333) State 1 (granular ridges on the corpus bursae apex) is autapomorphic for Eosphoropteryx.
Sclerotized ridges (state 2) either unite the members of the Autoplusiina, with subsequent loss in
Erythroplusia, or are convergently derived in Rachiplusia and clade 31. Within the latter, very large,
strong folds have either arisen in clade 31, only to be lost independently in Loboplusia and
Allagrapha, or have been convergently acquired by clades 33 and 35. Only further examination of
additional taxa will resolve these ambiguities but character 213 is undoubtedly important in assessing
the interrelationships of the autoplusiine taxa. State 2 has also been convergently derived in Diloba,
Chalcopasta, Omorphina and Plusiopalpa.

214 (0-250) A recurved apex of the corpus bursae has been convergently derived in Pseudoplusia and
clade 14. However, reversal is interpreted to have occurred in Acanthoplusia and Stigmoplusia.

215 (0-429) This character was discussed in detail above in connection with the positions of Plusia,
Cornutiplusia and Autographa. State 1 (a patch of thickened membrane on the RHS corpus bursae) is
apomorphic for clade 40 but with a reversal in clade 44. State 2 (a shallow pouch) unites Autographa
bractea and Cornutiplusia, while state 3 (a large pouch or appendix) is autapomorphic for the latter
genus. State 3 arises de novo from state 0 in Calophasia.

Miscellaneous
216 (0-500) Autoplusia and Rachiplusia are interpreted as being convergently sexually dimorphic.

Clade analysis

Each clade, as numbered in Figs 27, 28, is followed by its name (if applicable) and a list of its apomorphies
in square brackets. The states printed in italics are those that are considered to be unique and unreversed
synapomorphies. For those characters in which the state assignments to the internal nodes are ambiguous,
the optimizations described in the preceding section are employed. It should be remembered, however,
that these do represent only one of many possible interpretations.

Clade 1: Plusiinae [15(1), 27(1), 29(1), 31(1), 32(0), 33(0), 35(0), 49(1), 131(1), 143(3), 782(7), 184(1)].
The Plusiinae have been conventionally defined as the lashed-eyed, quadrifine noctuids. This study has
demonstrated that this characterization is inadequate. Many of the outgroup taxa have lashed eyes
(although it must be admitted that several were chosen for inclusion in this investigation expressly because
they displayed this state), while one plusiine genus (Mouralia) has a distinctly trifine hindwing venation.
Detailed study of the condition of hindwing vein M, showed that there are various degrees of reduction (see

SYSTEMATICS OF THE PLUSIINAE 139

character 15) and that a straightforward division into present and reduced/absent is too simplistic. Thus, a
new characterization of the subfamily must be formulated.

Of the 11 character states listed above, four are unique and unreversed for all members of the Plusiinae
and support the monophyly of the subfamily. Thus, the Plusiinae can be defined, on present evidence, as
those noctuids with a convex occiput (character 29); few apical styloconic sensilla on the proboscis (31),
which has strengthening bars that are semicircular on the most apical portion (35); and anal papillae that
bear differentiated, dorso-ventral rows of setae (182).

With the possible exception of the last, however, these characters are not particularly useful as key
characters. Several of the other seven homoplasious characters listed above are more suited to this role,
providing that the exceptions are recognized and noted. Their inclusion results in a polythetic, but
nevertheless workable, definition of the Plusiinae. The other characters are of three types. First, and most
important, is character 184, the apomorphic state of which (the presence of an eversible dorsal pouch
between the anal papillae and A8) is interpreted as characterizing the Plusiinae. Although it appears to
have been lost in the Abrostolini, this pouch has not yet been found in any other group of noctuids and
constitutes additional evidence corroborating the monophyly of the Plusiinae. Second, there are those in
which the apomorphic state is found in all the plusiines but also occurs in one or more of the outgroup taxa:
triangular pilifers (27; also found in Calophasia); a long, thin proboscis (32; also found in Calophasia +
Cucullia), with smooth styloconic sensilla (33; also occurring in Cucullia); and a large thoracic crest (49;
present also in Chalcopasta). Third, the apomorphic state may not only occur in the outgroups, but may
also be transformed within the Plusiinae into other states, which themselves are found in the outgroups or
are homoplasious within the Plusiinae. These are: a hindwing vein M,; that is half as strong as M; (15; this
state also occurs in Diloba, while within the plusiines, it is transformed into state 2 [a vestigial vein M)] in
Mouralia); the presence of saccular flanges (131; these are lost in the Abrostolini, Trichoplusia and
Autoplusia); and a clavus borne on a peninsula on the sacculus (143; this condition is further modified in the
Abrostolini and Anagrapha into state 2, in which the dorsal and ventral margins of the clavus are basally
more or less level, convergent upon the condition found in Cucullia).

Thus, it would appear that the key characters commonly used in the past are generally either too
homoplasious (have too many exceptions to be of use to the non-specialist) or, on close examination, prove
to have been misinterpreted. It is becoming increasingly apparent that satisfactory diagnoses of higher-
level taxa will in future rely increasingly heavily upon character states that are not easily observable. This is
a price that must be paid for a predictive and useful higher-level classification.

The subfamily Plusiinae is represented throughout the World, with the exception of Antarctica and
perhaps some of the more remote oceanic islands. A large number of species occur in Africa and the
eastern Palaearctic, while, in comparison, the fauna of South America is relatively depauperate. Plusiines
appear to be largely absent from tropical lowland forest habitats, while they reach their greatest diversity in
tropical montane and temperate areas. Twelve species occur in Great Britain, either as residents or regular
immigrants (e.g. Autographa gamma), while rare immigrants (e.g. Thysanoplusia orichalcea), singleton
occurrences (e.g. Ctenoplusia accentifera Lefevbre) and doubtful natives (Euchalcia variabilis) account for
a further twelve species. They rejoice under such common names as the Spectacle (Abrostola triplasia), the
Beautiful Golden Y (Autographa pulchrina (Haworth)), the Burnished Brass (Diachrysia chrysitis) and
the Gold Spot (Plusia festucae). Elsewhere in the World, the common names of plusiines are less
imaginative, if perhaps more practical; for example, le Gamma (Autographa gamma), the Bean Leaf
Skeletonizer (Autoplusia egena) and the Soybean Looper (Pseudoplusia includens).

Few generalizations can be made regarding plusiine biology. The adults fly strongly and a number of
species are migratory. However, they do not appear to be particularly attracted to light (Warnecke, 1953;
Moucha & Novak, 1956; Eichlin & Cunningham, 1978). Thus a light-trap is neither an efficient nor
accurate method for assessing plusiine population densities. Eichlin & Cunningham (1978) related an
anecdotal observation that although numerous plusiines were observed feeding at night from flowers of a
Button Bush (Cephalanthus occidentalis) that was growing less than 15 feet from a black light trap, only two
moths actually came to the light. Many plusiines fly equally by day and by night. Silver Ys (Autographa
gamma) are frequently seen in Britain on sunny late summer days feeding from Michaelmas Daisies (Aster
spp.) in the company of butterflies and bees.

As far as is known, temperate region species are univoltine, while in warmer climates, several
generations per year seem likely. However, it should be stressed that only a minority of life histories,
particularly of tropical species, has so far been elucidated. Multiple matings appear to be the rule, and the
eggs are generally laid singly on the larval foodplant, probably between 200 and 300 per female (Eichlin &
Cunningham, 1978). Plusiine larvae are usually green with white, often interrupted, longitudinal stripes.
The head may bear a black, lateral stripe. Several species are sensitive to population density, becoming
darker when overcrowded (Ichinosé, 1959). There are generally five larval instars but a few species pass

140 IAN J. KITCHING

through six or even seven. In most species, the prolegs on segments A3 and A4 are reduced or missing (see
below) and the larvae proceed by a looping action, thus giving rise to their common name of semiloopers.
Pupation is usually in a cocoon spun between leaves. Pupal coloration has been shown to be temperature
dependent in at least one species (Ichinosé & Asawa, 1964) with darker pupae being formed at lower
temperatures. According to Eichlin & Cunningham (1978), the prepupal stage lasts 1-2 days and the adults
emerge about eight days after pupation. Little is known about plusiine courtship behaviour, but judging
from the widespread occurrence of various tufts, brushes and pouches, pheromones produced by both
sexes are likely to play an important role (see Gothilf & Shorey, 1976, for a comprehensive account of
courtship in Trichoplusia ni).

Omorphinini [7(1), 20(0), 24(1), 42(1), 52(7), 59(1), 127(1), 137(1), 144(1), 147(1), 213(2)]. This tribe
consists of a single genus and species, Omorphina aurantiaca. Of the character states listed above, ten are
homoplasies occurring elsewhere in the Plusiinae or among the outgroups. This is only to be expected since
the genus-level autapomorphies were largely eliminated from the data set prior to the analysis. However,
characters 7 (yellow hindwings), 20 (randomly arranged scale bases), 24 (a prominent rim around the eye)
and 137/147 in combination (oval valves with setae arising from marginal prominences) are useful
recognition features. In addition, state 52(1), a truncate dorsal tegular arm, is autapomorphic. Although
considered by Kostrowicki (1961) to be more correctly treated as an acontiine, the presence in Omorphina
of the character states listed above for clade 1 argue strongly that Hampson (1913) was correct in placing
the genus in the Plusiinae.

Omorphina aurantiaca (Fig. 31) is a small moth, recorded only from Xizang Province, China (Tibet) and
Sikkim, at around 15,000 ft (4500 m). Nothing is known about its biology except that the adult flies in July
(Alphéraky, 1892). Omorphina has previously been likened to the Palaearctic amphipyrine Panemeria
tenebrata (Scopoli) (Alphéraky, 1892), but apart from the presence of marginal prominences on the valves,
the genitalia of the two genera are dissimilar in both sexes. In particular, Panemeria lacks the row of setae
on the anal papillae, while the cornuti on the vesica are very similar to those of Chalcopasta, which may
indicate that the true affinities of Panemeria may lie with the Stiriini.

Clade 2 [10(1), 28(0), 58(1), 73(1), 80(1), 99(1)]. This clade, which comprises all the Plusiinae except
Omorphina (Omorphinini), is well supported by four unique and unreversed apomorphies: a dorsally
delimited frons (28); an internal flange on the anterior edge of T2 that is produced subdorsally (73; see also
clade 4); a particular form of St8 and ventral hair pencil (80), although this is frequently further modified;
and the presence of a notch between the tegumen and uncus (99). In addition, all members of clade 2
display a metallic sheen over much of the forewings, although this is often weak and difficult to observe (10;
the condition in Chalcopasta is almost certainly convergent but needs to be checked in other members of
the Stiriini), and often bear a white spot on the fore tibia (58; this is frequently lost in the Plusiinae, occurs
also in Chalcopasta and the Eutelinae, and is often impossible to score in many taxa due to additional
pattern modifications).

Clade 3: Abrostolini [1(1), 2(7), 51(1), 100(1), 102(1), 103(1), 107(1), 124(1), 125(1), 126(1), 131(0),
134(1), 143(2), 169(1), 170(1), 184(0), 187(1), 188(1)]. There can be no doubt that the abrostoline genera
Abrostola and Mouralia form a monophyletic group. Of the 18 character states listed above, no fewer than
13 are unique and unreversed. Members of the tribe are best recognized by the form of the forewing
stigmata (2), the fusion of the aedeagus to the juxta apex (124), the spatulate saccular lobe (134) and the
desclerotized band and invaginated pocket laterally on T8 in the females (187/188 respectively). In
addition, they are the only plusiines to possess raised scales in the forewing cell (1) and to lack an eversible
pouch between the anal papillae and A8 in the females (184). Uniquely in the subfamily, as far as is known,
the larvae of the Abrostolini bear prolegs, albeit somewhat reduced, on abdominal segments 3 and 4
(Eichlin & Cunningham, 1978; see also below), although this condition is probably plesiomorphic.

Abrostola [22(2), 121(2), 143(1), 169(2)]. This genus is probably paraphyletic relative to its sister-taxon
Mouralia. The ventral hook on the vesica (169) and the heavily lashed eyes (22) appear to be restricted to
A. triplasia (Fig. 32) and a few other, possibly closely related, species (e.g. A. kaszabi Dufay [Dufay, 1971];
A. trigemina (Werneberg)). The blunt, lobate VA VB (121) and the broad, rounded clavus (143) need to be
confirmed in the other species of Abrostola.

Abrostola includes about 35 species, of which four are North American and the remainder almost
equally divided between Africa and the Palaearctic. None are known from South America, South East
Asia or Australasia. Many species have only recently been described, particularly important contributions
in this respect being those of Dufay (1957; 1958; 1960; 1971). The larvae of Abrostola are characteristic in

SYSTEMATICS OF THE PLUSIINAE 141

having a ‘humped’ segment A8, usually bearing two small, subdorsal points. In most species (e.g. A.
triplasia, A. trigemina, A. major Dufay), there are diagonal, subdorsal white or cream stripes on segments
A1-8, together with darker, more dorsal markings. On segments Al, A2 and A8, the latter fuse to form
single, dorsal spots, which, together with the more pronounced pale markings, produce a very distinctive,
somewhat eye-like, pattern. The overall colour can vary from green to violet-brown. The larva of A.
asclepiadis is very different. A single, dried, inflated skin in the BMNH collection is uniform pale cream
(?green in life), and is covered in a pattern of small, black spots, not all of which are associated with setal
pinacula. The pattern is suggestive of an aposematic colour scheme, which is not unreasonable given the
species’ asclepiadaceous foodplant (see below). However, this hypothesis remains to be corroborated by
the description or a colour photograph of a living larva. The biology of only a few species is known,
primarily from the Holarctic fauna. Most species for which the larval hostplant is known feed on
Urticaceae; usually Urtica itself (Crumb, 1956; Beck, 1960; Ichinosé, 1962b; Bretherton er al., 1983;
Miyata, 1983), but A. anophioides Moore utilizes Girardinia heterophylla (Gardner, 1947), while A.
trigemina and A. major also feed on Boehmeria. A. abrostolina (Butler) is recorded from Fatoua
(Moraceae) (Miyata, 1983), and other recorded foodplants include Epilobium, Lamium (Hampson, 1913)
and Humulus (Bretherton et al., 1983). One species is exceptional, however, in that it has performed a
remarkable hostplant shift; A. asclepiadis is the only plusiine known to feed exclusively on an asclepiad,
Vincetoxicum hirundinaria (Hampson, 1913 as Cynanchum vincetoxicum, Beck, 1960; Weidlich, 1984 as
Vincetoxicum officinale). Members of Abrostola are also unusual among the plusiines for overwintering in
the pupal stage rather than as early instar larvae (Eichlin & Cunningham, 1978). The British species
are known as ‘spectacles’ from the markings on the patagia, but this pattern is not found throughout the
genus.

Mouralia[14(2), 15(2), 18(2), 19(1), 64(3), 83(5S), 136(1), 191(1)]. Mouralia currently includes only a single
described species, M. tinctoides (Fig. 33). Of the above eight listed character states, three constitute good
autapomorphies for the genus: the presence of spines on the inner surface of the hind tibia (64); the
extremely long sternal hair pencil, arising from a single pocket that almost reaches the base of the abdomen
when retracted (83); and the presence of basally-directed scales on the valve apex (136). Although this
latter condition was coded as also present in the Argyrogrammatini (q.v.), the form of the scales differs and
so the two conditions probably represent distinct characters. In the Argyrogrammatini, the scales are
simple and setose, often quite stiff and usually easily detached, while in Mouralia. these scales are long,
soft, curved, apically spatulate and persistent (Fig. 348). In addition, Mouralia is the only plusiine with a
trifine hindwing venation (15) and has a female frenulum reduced to two setae (14; a condition also found in
clade 6, Rachiplusia, Autoplusia and Brachionycha).

There are also numerous abdominal and genitalic autapomorphies that were not included in the analysis.
These include: a long, thin, medial apodeme projecting anteriorly from the anterior margin of St7 in the
male (this may be functionally correlated with the long hair pencil pocket); an apically, laterally flattened
uncus (Fig. 278); a pair of spatulate flaps arising from the dorsal sclerotized band on the anal tube; an
ampulla (Fig. 348); valves thickly covered ventrally and exteriorly with long, flexuous, pale yellow, ‘sticky’
(in the sense that they adhere to forceps) scales; valves in which the apices turn through 90 degrees so that
the costal edges become internal and the saccular edges become external (Fig. 348); and a corpus bursae
with a very elongate, tapering fundus that broadens to a blunt tip.

A second undescribed species of Mouralia has recently been discovered in Bolivia (F. Navarro, pers.
comm. ). It differs in several features from M. tinctoides. The most notable of these are (as assessed from a
photographic slide of a male genitalic preparation) the presence of only a rudimentary hair pencil and a
valve apex that lacks all the modifications of M. tinctoides, simply tapering to a point as in many abrostolas.
Thus this new species may represent an intermediate form between the genus Abrostola and Mouralia
tinctoides.

Another candidate for such an intermediate taxon is Abrostola anophioides. This is a large species found
in the Himalayan foothills from Uttar Pradesh in NW. India to Sichuan Province, China. Unlike other
_ abrostolas, the males of A. anophioides have a short, single hair pencil, while the fundus of the corpus
bursae is elongate and apically swollen as in Mouralia tinctoides. A. anophioides and the new Mouralia
species strongly suggest that one result of a detailed study of the Abrostolini might be the synonymy of
Mouralia with Abrostola.

Mouralia tinctoides is widespread throughout the Neotropics, penetrating as far north as the SW. United
States. Its life history has been described in detail by Comstock (1937). The general appearance of the larva
is that of an Abrostola but it differs markedly in hostplant, feeding on Tradescantia fluminensis and Zebrina
pendula (Commelinaceae) (Eichlin & Cunningham, 1978) (although the latter may be an error; Eichlin,
1975). Unusually for the Plusiinae, there are seven larval instars (the norm is five).

142 IAN J. KITCHING

Clade 4 [3(1), 4(1), 6(1), 59(1), 73(2), 83(3), 106(1), 157(2), 158(1), 162(1)]. The Argyrogrammatini and
the Plusiini are united as sister-groups primarily on the basis of two synapomorphies: a distinct, bulbous
coecum (157) and a unique form of the internal flange on the anterior edge of T2, in which the two
subdorsal lobes (see clade 2) are fused medially but with a medial ventral emargination (73). In addition,
several character states occur only in clade 4, although in not all of the constituent genera; these are the
characteristic plusiine Y-mark (3), a pink terminal line on the forewing (4, see also clade 25), a scale tooth
(6), restricted sclerotization on the aedeagus (158) and a subapical cornutus (162). Further, in all members
of the Argyrogrammatini and Plusiini, the LASTs are fused around the base of the uncus, although this
condition recurs in Chalcopasta.

As far as is known, all members of clade 4 that hibernate do so as early instar larvae. Also, the prolegs are
vestigial or absent on segments A3 and A4 in all larval instars (but see Anadevidia).

Clade 5: Argyrogrammatini [14(1), 18(0), 89(1), 108(1), 136(1), 144(1), 158(2), 173(1), 209(1)]. This
primarily tropical tribe of plusiines is quite well characterized by three synapomorphies. All argyrogram-
matines have secondary hair brushes arising on St8 (89), basally-directed setose scales apically on the valve
(136, see also Mouralia), and a differentially sclerotized band on the vesica (173). Seta 1 of the female
frenulum is also reduced (14; see Mouralia, Autoplusia, Rachiplusia) and the ventral sclerotization of the
aedeagus is further restricted to a distinct narrow band (158; with convergence occurring in the autoplu-
siine genera Erythroplusia and Macdunnoughia).

Anadevidia [3(0), 6(1), 105(1), 116(1), 148(0)]. Anadevidia is an enigmatic genus, of somewhat uncertain
position. Both Kostrowicki (1961) and Ichinosé (19625) refrained from making any explicit statements as
to its relationships, although the latter derived the larval setal pattern from that of Thysanoplusia (as
Diachrysia), a relationship formalized by Chou & Lu (1979) as the ‘Plusiini’. Kljutshko (19855) placed
Anadevidia in the Plusiini, where it remained after the reanalysis performed above, on the basis of lack of
prolegs in the mature larva and setae SD1 on segment A9 being ‘normal’. This similarity to members of the
Plusiini is supported further by character 116, in that both taxa possess a strongly concave pleurite.
However, as described above, there is stronger evidence (characters 89, 136 and 173) to suggest that
Anadevidia correctly belongs in the Argyrogrammatini.

It may yet be shown that Anadevidia is in some way related to Chrysodeixis (+ Pseudoplusia). Both
genera bear spatulate scales on segment A8 in the male (although these are interpreted here as
independent conditions). Also, the vincular arms in Anadevidia are long and narrow, and superficially
resemble those of Chrysodeixis and Pseudoplusia, although they are separate in the former genus and
intimately fused into a thin spine in the latter two.

Anadevidia currently contains only two species, A. peponis (Fig. 34) and A. hebetata. The former is
widespread throughout the Indo-Australian region, while the latter is apparently restricted to India and
Japan. In general appearance, they are dull grey-brown moths with no distinguishing pattern features, and
bear a superficial resemblance to species of the genera Allagrapha and Diachrysia (especially D. leonina
and D. bieti (Oberthir)). The larvae of Anadevidia have several unusual features. They are the only
plusiines in which the first instar larvae possess on segments A3 and A4 well developed prolegs that are
subsequently lost completely in the later instars. The larvae also bear enlarged pinacula on some segments,
giving them a warty appearance. Anadevidia is oligophagous, feeding on members of the Cucurbitaceae,
although A. peponis has been recorded from Mercurialis (Euphorbiaceae) and Paulownia (Scrophular-
iaceae) (Miyata, 1983). The larval chaetotaxy of Anadevidia has been described by Ichinosé (1958);
1962a; b).

Clade 6 [12(1), 14(2), 76(1), 77(1), 83(2), 91(1), 94(2), 110(1), 167(1), 185(1)|. Three very good
synapomorphies characterize clade 6: the form of St7 in the male (76); the presence of an invaginated sac in
the vesica (167); and the presence of a sub-basal dorsal flange on the posterior apophysis (185). The males
of the majority of the constituent taxa also bear dorso-lateral hair tufts on at least segments AS and A6 (77,
except Zonoplusia and Pseudoplusia), have dorsally twisted lateral arms of St8 (91, except Pseudoplusia)
and have broad, ‘greasy’ scales apically on the dorsal arms of St8 (94, again, except Pseudoplusia). Overall,
clade 6 is very well supported, allowing for the multiple reversals of Pseudoplusia (q.v.).

Clade 7 [93(1), 105(1), 108(0), 128(1), 162(0), 172(1), 174(1)]. The three genera of this clade are united bya
single synapomorphy, a dorsal, convex, rounded plate on the vesica basally (174). In addition, a pair of
short, straight cornuti occur in both Dactyloplusia and Pseudoplusia, but not in Chrysodeixis (172).
However, as discussed above (Character analysis), the condition in the latter genus may prove to be
derivable from that in the former two. Also, all three genera have a juxta that is apically scobinate (128),
although this condition recurs in Thysanoplusia, Plusiopalpa and Cucullia.

a a a ae ase

Ree eS TS 2 ers

ey a ae

SYSTEMATICS OF THE PLUSIINAE 143

Dactyloplusia [60(1), 78(1), 80(2), 92(1), 110(0), 149(1)]. In the present analysis, Dactyloplusia is
interpreted as being the sister-group of Chrysodeixis + Pseudoplusia, on the basis of the characters listed
above for clade 7. However, there are an almost equal number of character states that suggest Dactyloplu-
sia may instead be related to clade 10, and in particular, Thysanoplusia. These include the presence of a
hind basitarsal comb (60), latero-ventral flaps on the anterior bar of St8 (80), a sub-basal lobe on the lateral
arms of St8 (92), a sharply hooked harpe (149), as well as the apically scobinate juxta referred to in the
preceding paragraph. Dactyloplusia is also autapomorphic in having the dorso-lateral hair tufts on segment
7 enclosed in a groove (78).

Dactyloplusia currently contains only a single species, D. impulsa (Fig. 50). Holloway (1985) gives the
distribution as Sri Lanka, India, S. China, Sundaland, Sulawesi and New Guinea. However, by far the
majority of specimens in the BMNH collection were caught in Sri Lanka. Nothing is known of the biology
of D. impulsa. Holloway (1985) captured a Bornean specimen ‘on a river bank with disturbed vegetation
and alluvial forest near a limestone cliff at 100 m’.

Clade 8 [11(1), 771(1), 120(1), 130(0), 144(0), 196(1)]. A sister-group relationship between Pseudoplusia
and Chrysodeixis is hypothesised on the basis of two synapomorphies: a long, thin, spine-like vinculum
(111; the condition found in Stigmoplusia is considered to be non-homologous) and a very narrow, almost
straight pleurite (120). Ichinosé (1973) also recognized the great similarity between these two genera and
thought that Pseudoplusia might eventually prove to be best placed in Chrysodeixis, perhaps as a subgenus.

Pseudoplusia [59(0), 77(0), 83(0), 91(0), 93(0), 94(0), 728(2), 209(0), 214(1)]. Pseudoplusia is perhaps
most notable for the total reduction of the male scent organs, both those of St8 and the dorso-lateral hair
tufts (characters 77, 83, 91, 93, 94). This may prove to be due to the geographic isolation (one species in the
Americas, the other on Saint Helena) from other members of clade 7, resulting in a relaxation of the
selective pressures that maintained the elaborate scent organs, which were thus lost. Pseudoplusia also
displays a number of autapomorphies. The males possess a hook-like process near the base of the valve
costa, an extremely long, thin, sinuous harpe, and bear a sharp spine medially on the posterior margin of
the juxta (Fig. 341; similar to the type of spine found in some species of Syngrapha). The juxta is also
latero-apically spinose (128). In the females, the apex of the corpus bursae is extremely elongate and
tapered, projecting first anteriorly and then recurving posteriorly.

The genus Pseudoplusia contains two species, P. includens (Fig. 38) and P. dalei (Wollaston) comb. n.
(see Appendix 4). The former is widespread from Canada to Argentina and Chile (Forbes, 1954; Angulo &
Weigart, 1975) and is an important pest of many cultivated plants (Canerday & Arant, 1967). It is
commonly known as the Soybean Looper (Eichlin & Cunningham, 1978) or, less frequently, the False
Cabbage Looper (Canerday & Arant, 1967). An icosahedral virus has recently been isolated from P.
includens (Chao, Young & Kim, 1984) and may prove useful as a biological control agent. P. dalei is
endemic to the mid-Atlantic island of Saint Helena. Little is known of its biology except that the adults
have been seen frequenting the flowers of geraniums and Pittosporum and that it may belong to the
nearly-extinct scrubland fauna of the island (Wollaston, 1879).

Chrysodeixis [12(0), 77(3), 83(5), 94(1), 97(1), 105(0), 127(1), 172(0), 207(1)]..In contrast to its sister-
genus, Chrysodeixis is most notable for the often extreme elaboration of the scent organs on the male
abdomen. The progressive reduction of the sternal hair pencils, with the concomitant modification of the
secondary hair brushes, T8 and St8, has already been described above (Terminology). In addition, the
more derived species, such as C. chalcites (Fig. 39), have distinct additional dorso-lateral hair tufts on
segment A4 (77); broad, spatulate scales apically on the dorsal arms of St8 (94); and an additional two pairs
of hair pencils arising subventrally from discrete pockets in the membrane between segments 7 and 8. In the

more plesiomorphic species, C. argentifera (Fig. 40), the sternal hair pencils are more fully developed but

arise from a single pocket (83), much as in Mouralia.

Chrysodeixis is a moderate-sized genus of about 25 species found throughout the Old World tropics.
However, by far the majority of species occur in SE. Asia and Australasia. Most are relatively restricted in
Tange but three, presumably by virtue of a strong migratory habit, have become very widespread. C. acuta
is found more or less throughout the entire range of the genus (Dufay, 1970a; Pinhey, 1975). It has even
been recorded from Great Britain despite apparently never having been captured on mainland Europe
(Bretherton er al. , 1983). C. chalcites is sympatric with C. acuta over the western part of the latter’s range,
but is replaced by C. eriosoma from approximately Indonesia eastwards. There is some debate as to
whether C. chalcites and C. eriosoma are (Pinhey, 1975) or are not (Kostrowicki, 1961; Dufay, 1970a)
conspecific. The biology of the two species is very similar and although allopatric C. eriosoma can be
distinguished from C. chalcites by differences in the cornuti (Kostrowicki, 1961), in the area where the two

144 IAN J. KITCHING

species are thought to overlap, the distinction is not as clear cut and there is confusion as to their identities
(Holloway, 1985). C. chalcites has also been taken in Great Britain and, in contrast to C. acuta, is
frequently recorded from Mediterranean Europe (Bretherton et al. , 1983).

One section of Chrysodeixis, the minutus Dufay species-group (Holloway, 1985), does not show the
needle-like vinculum but does possess additional hair pencils between segments A7 and A8. The
relationships of these species need to be assessed carefully in any future analysis of the interrelationships of
the members of clade 8.

Clade 9 [59(0), 203(1), 204(1)]. This clade is not supported by any good synapomorphies and will probably
not be corroborated by future studies. Two of the character states, an elongate ductus bursae (203) anda
transversely orientated corpus bursae (204), only occur in the included genera Zonoplusia, Thysanoplusia,
Trichoplusia and Eutheiaplusia (with Plusiotricha unknown), with both states being subsequently lost in
clade 14. However, the apomorphic states for 203 and 204 are not found throughout Trichoplusia and
Thysanoplusia (Dufay, 1970a) (Zonoplusia and Eutheiaplusia are treated here as monotypic) and it is thus
highly unlikely that they will be shown in future to support a monophyletic group of genera comprising
clade 9 less clade 14. Certainly, before such a study can be attempted, a critical reappraisal of Trichoplusia
and Thysanoplusia must be undertaken (see also Plusiotricha).

Zonoplusia [4(0), 58(0), 77(0)]. Zonoplusia is treated here as a monotypic genus, comprising only the type
species, Z. ochreata. When originally erected (Chou & Lu, 1978a), five species were placed in Zonoplusia.
Two were unavailable for examination (Z. longisigna (Chou & Lu) and Z. brevistriata (Chou & Lu)) but
are probably closely related to, if not synonymous with, either of the other two species. These are
albostriata, a current member of Crenoplusia (for the moment, see 60 and 133) and daubei, presently placed
in Thysanoplusia. Ichinosé (1973) included Z. ochreata in Thysanoplusia, although it differed from T.
intermixta and T. daubei in the larvae having very fine cuticular ornamentation (like Chrysodeixis,
Trichoplusia and Ctenoplusia s.\.) rather than large and hair-like microtrichia.

Z. ochreata (Fig. 49) is a small moth, in which the Y-mark has been modified into an oblique, pale grey,
double line. The subapical cornutus is unusual in that it arises from a sclerotized base that is almost as long
as the cornutus, thus giving the appearance of a double cornutus. Z. ochreata is found in the eastern
Palaearctic and throughout the Oriental tropics and has also been recorded from Queensland, Australia
(although otherwise not east of Sundaland and the Philippines (Kostrowicki, 1961; Holloway, 1985)). The
only known larval foodplant is Polygonum thunbergii (Polygonaceae) (Miyata, 1983).

Clade 10 [80(2), 84(1), 92(1), 93(2), 209(0)]. This clade is also not supported by any good synapomorphies.
As was discussed above, states 80(2) and 92(1) are shared with Dactyloplusia, which may have its true
affinities with members of clade 10. Sclerotized bars on the sternal hair pencil pockets (84) are restricted to
members of this clade but are absent in Thysanoplusia, Plusiopalpa and Acanthoplusia. Similarly, dorsal
arms of St8 that are apically fused into a rounded point (93:2) are found only in clade 10, although this state
is further modified in clade 18 (93:3) and Thysanoplusia (93:1, convergent upon the condition found in
Chrysodeixis and Dactyloplusia).

Clade 11: Trichoplusia s.1. [60(1), 83(3), 144(0)]. No unique and unreversed synapomorphies have been
found to link Trichoplusia and Thysanoplusia as sister-groups. The form of the hind basitarsal comb (60),
sternal hair pencils (83) and clavus (144) are all probably the plesiomorphic conditions for these characters
in the Plusiinae (whether by primary symplesiomorphy or by secondary reversal).

Trichoplusia s.str. [105(1), 110(0), 131(0), 149(1), 203(2), 205(1)]. Trichoplusia s.str. (i.e. sensu McDun-
nough, 1944; Ichinosé, 1973) is an extremely heterogeneous genus and appears to have been adopted as the
‘dustbin’ of the Argyrogrammatini, in much the same way as ‘Plusia’ s.1. accommodates any generally
unplaceable plusiine. As a result, the genus currently contains about 50 species. The remarks made below
thus properly refer only to T. ni.

The only good autapomorphy for T. ni (Fig. 43) discovered in this study was the extremely long ductus
bursae (203), which was the main character state emphasised by McDunnough (1944) when he originally
proposed the genus. A similar long ductus bursae also occurs, however, in T. oxygramma (McDunnough,
1944; Eichlin & Cunningham, 1978) and in T. gorilla (Holland) (Dufay, 1970a), where it is even longer,
being some 3-4 times the length of the body. T. ni is the only argyrogrammatine examined to lack a
saccular flange (131) and, like Dactyloplusia and Agrapha, possesses a sharp, hooked harpe (149).

Trichoplusia is badly in need of a critical and thorough revision. Species have been added somewhat
casually since Dufay (1970a) began the rapid expansion of the genus. It is almost certainly polyphyletic with

rae

ea

SYSTEMATICS OF THE PLUSIINAE 145

regard to at least Thysanoplusia, Dactyloplusia, Zonoplusia and Ctenoplusia. This much is immediately
obvious from the length of the couplet (24 lines) necessary to distinguish Trichoplusia from Ctenoplusia in
Dufay’s (1970a) key. A flavour for the problem can be gained from the description in the key of the form of
the cornuti in Trichoplusia (translated freely): ‘aedeagus often armed with a small, spiniform or needle-like
cornutus, accompanied or not by a mass of fine spicules, or furnished with a large number of spines, or only
with a more or less dense edging of denticles; rarely almost unadorned’. The corresponding ‘condition’ in
Ctenoplusia reads: ‘aedeagus armed with many cornuti or with a large and long, stick-like cornutus, never
with a single, fine, spiniform cornutus, sometimes with several small cornuti’. Trichoplusia almost certainly
contains a number of monophyletic units and their elucidation is probably the most pressing, yet most
difficult, task in plusiine systematics.

T. ni is the world’s most widespread plusiine. As race brassicae (Riley), it occurs from Canada to
Central, and even South, America (Kostrowicki, 1961; Eichlin & Cunningham, 1978) and had been
introduced into Hawaii by 1939 (Zimmerman, 1958). As race ni, the species occurs throughout the
southern Palaearctic, Africa and most of the Oriental and Australasian regions, reaching South Africa,
Madagascar, Korea and New Zealand. The two races are very dubiously distinct (the forewings of ni are
said to be lighter and often greyer than those of brassicae) and the separation is often not made (e.g. Eichlin
& Cunningham, 1978; Franclemont & Todd, 1983).

Commonly referred to as the Cabbage Looper or the Ni Moth, T. niis a very important pest of numerous
crop plants, but particularly of brassicas (as one of the common names implies). The list of its larval
foodplants is impressive. Eichlin & Cunningham (1978) list over 60 genera from a wide range of families
and the species would appear to be able to eat virtually anything, including plants with apparently
considerable chemical defences (e.g. Asclepias, Crotalaria, Euphorbia and Senecio). I reared large
numbers, from New Zealand stock, in the BMNH and the only plant that was actively refused was the
introduced Japanese knotweed, Reynoutria japonica. Because of its pest status, 7. ni has been the subject
of much research activity aimed at its control or destruction and this in turn has generated an enormous
amount of literature on the subject (see any appropriate abstracting journal).

Thysanoplusia (2(0), 9(1), 84(0), 93(1), 112(1), 128(1), 159(1)]. Thysanoplusia was originally erected by
Ichinosé (1973) to accommodate those species previously incorrectly placed in Diachrysia, due to
confusion over type species (see Review). The genus originally included T. intermixta (Fig. 47), T.
orichalcea, T. daubei (Fig. 48) and T. (now Zonoplusia) ochreata. Suggested future candidates for
inclusion were florina, homoia (Dufay), ignicollis (Dufay), semirosea (Walker) and viettei. Although
stated to be separable on the basis of larval and adult characters, the latter were never clearly stated. Thus,
the characterization of Thysanoplusia came to depend upon larval features. Most notably, Thysanoplusia
differed in having a very slender seta SD1 on segment A9 and in bearing setae SV1 and SV2 on A2 ona
single pinaculum (Ichinosé, 1962b; 1973). The genus was retained, with reservations, by Holloway (1985),
who added T. ekekei (Bethune-Baker) and T. bipartita (Snellen). However, most authors treated
Thysanoplusia as synonymous with Trichoplusia (e.g. Dufay, 1975; 1982b) and eventually it was so treated
in the Japanese checklist (Inoue et al., 1982). Although I accept that Thysanoplusia defined in terms of the
two larval characters may prove to be a monophyletic unit, until such time as good adult synapomorphies
are found (or, alternatively, and far less likely, the larvae of the outstanding species are described), the
concept is of limited value. Thysanoplusia is best either sunk completely into Trichoplusia until a
comprehensive revision can be undertaken, or restricted to a group of species centred upon T. intermixta
(perhaps including 7. daubei and its relatives), but always remembering that this is at best an incomplete
group.

T. intermixta is a member of that subsection of Thysanoplusia with a conspicuous metallic golden-green
patch on the forewing (9). Five species are currently recognized in this species-group: T. ekekei from New
Guinea; T. bipartita from Sulawesi and Borneo (both these species have diffuse patches); T. intermixta
from the Oriental tropics as far as Java and Japan (Holloway, 1985); 7. florina, a Madagascan endemic;
and T. orichalcea, which is widespread in the Old World tropics and subtropics. Like other highly
successful and widespread plusiines (Trichoplusia ni, Chrysodeixis acuta, C. chalcites, Pseudoplusia
includens, Autographa gamma), Thysanoplusia orichalcea is strongly migratory. It is occasionally taken in
Great Britain and Scandinavia and has recently become established in New Zealand (1984), possibly from
New Caledonia or Australia. Again, like the other successful species, T. orichalcea is a serious agricultural
pest throughout much of its range, including New Zealand (J. Dugdale, pers. comm.), where it attained
this status within a mere 12 months of its recorded arrival.

Clade 12 [12(0), 77(2), 91(2), 138(1), 162(0)]. This clade is another in the Argyrogrammatini that is not
corroborated by any good synapomorphies. Two features are restricted to members of this clade: the

146 IAN J. KITCHING

presence of weak dorso-lateral hair tufts on segments (A2—)A4 in the male (77) and lateral arms of St8 that
are twisted dorsally with a deep, basal concavity (91). However, both of these states are later lost, in clade
16 and Plusiopalpa respectively.

Clade 13 [108(0), 740(1), 159(1), 160(1), 205(1), 209(1)]. Eutheiaplusia and Plusiotricha are considered to
form a sister-group pair on the basis of two character states: a broad, blunt point on the sacculus (140) and
the characteristic serpentine form of the inverted vesica (160). In addition, the ornamentation of the vesica
is unusual in comprising more or less uniformly distributed small spines (159), a condition that is elsewhere
only found in Thysanoplusia.

Plusiotricha [11(1), 60(2), 61(1), 272(3), 119(1), 133(1), 144(2)]. As discussed above, Plusiotricha shares
with Ctenoplusia and Acanthoplusia the unique blade-like setae along the sacculus. In addition, Plusio-
tricha also shares another state with clade 16, a pleurite and VAVB fused for about half the length of the
former (119), and has a similar form of basitarsal comb (60) to Ctenoplusia. These character states argue
strongly that a future, more thorough study may show these three genera (and Eutheiaplusia, see below)
form a monophyletic unit. Plusiotricha is unique in having hook-like projections on the vinculum arms
towards their apices (112).

Of more immediately importance would be the discovery and examination of a female of Plusiotricha.
The genus currently contains three species, P. livida, P. carcassoni Dufay and P. fletcheri Dufay,
represented in collections by seven, one and ten males respectively. A further four males are known that
may represent a fourth species (Fig. 51). Female specimens are urgently needed to check whether the
predicted female character states, particularly the elongate ductus bursae (203), found in Eutheiaplusia do
also occur in Plusiotricha.

Nothing is known about the biology of Plusiotricha. The holotype of Plusia dyscapna Fletcher (a
synonym of Plusiotricha livida) includes a pupal exuvium and final instar larval skin, but no life history
information accompanies the specimen. Neither was any mention made of the immature stage material in
the original description (Fletcher, 1963). The genus Plusiotricha appears to live in mountainous areas at an
altitude of about 4000-5000 ft (1200-1500 m) in the tropical rainforest belt of Africa, from the Ivory Coast
to Kenya and Zimbabwe (Dufay, 1972). The largest number of specimens has been taken in Tanzania.

Eutheiaplusia [4(0), 83(0)]. Eutheiaplusia can be characterized as a Plusiotricha in which the male scent
organs have become highly reduced and presumably non-functional (83; and thus sits in relation to
Plusiotricha as Pseudoplusia does to Chrysodeixis), and in which the blade-like saccular setae are replaced
by fine, somewhat flattened, upcurved hairs and the vincular hooks are lacking. It could thus represent
either the sister-group of Plusiotricha or may eventually prove to be the most basal lineage of a clade
comprising the present clades 13 and 16. The vesica of the male illustrated by Dufay (1970a) bears a long,
thin, subapical cornutus absent in the specimen examined in this study. The presence or absence of this
structure in E. pratti (Fig. 52) requires confirmation.

Eutheiaplusia comprises a single species, E. pratti, endemic to Madagascar. Nothing is known of its
biology, although the species appears to be quite widespread on the island in forest at altitudes of between
730 and 1700 m (Dufay, 1970a).

Clade 14 [86(1), 87(1), 112(1), 203(0), 204(0), 214(1)]. As was found for clades 9, 10 and 12, there are no
good synapomorphies for clade 14. Accessory pouches in the sternal hair pencil pockets (86) occur only in
this clade (but not in Ctenoplusia or Plusiopalpa), as do sternal hair pencils in which the outer scales are
shorter than the inner (87; but not in the included clade 17). Dorsal emarginations of the vincular arms
(112) do characterize the members of this clade but have been convergently acquired by Thysanoplusia.

Agrapha [6(1), 12(1), 79(1), 88(1), 98(1), 105(1), 130(0), 144(2), 149(1), 162(1), 196(1), 207(1)]. Agrapha
currently contains three described species: A. ahenea, A. meretricia (Schaus) comb. n. and A. calceolaris
(Walker) comb. n. There are also two undescribed species that were allocated manuscript names by E. L.
Todd (note in BMNH collection, 1972). One, here termed species A, is represented in the BMNH
collection by seven specimens from Colombia, Bolivia and Peru, and includes noctuid genitalia prepara-
tion 5193 (male). The second species (sp. B) is unrepresented in the BMNH collection, is of unknown
provenance and depository, and was not examined.

Of the included species, three (A. ahenea (Fig. 57), A. meretricia and sp. A) present a distinctive facies,
particularly with regard to the highly developed dorso-lateral hair tufts, which give the abdomen a
truncate, bushy appearance. The same three species (A. calceolaris was not examined) have unique
flattened scales on the valve (see 133, Character analysis) and A. ahenea (and maybe A. meretricia and sp.

;
;

SYSTEMATICS OF THE PLUSIINAE 147

A) is the only argyrogrammatine to have discrete edges to the dorso-lateral hair tufts on both segments A5
and A6 (79).

Agrapha [ahenea] shares two character states with Argyrogramma (a distally dilated uncus, 98, and a
long inner tuft in the sternal hair pencil, 88) that must be interpreted as convergent. The hook-like harpe
(149), referred to above in connection with Dactyloplusia and Trichoplusia, is present in A. ahenea and A.
meretricia but in sp. A, the harpe is ‘normal’ (state 0) but with a small apical point. All species of Agrapha,
with the possible exception of A. calceolaris, have very long antennae (although this is difficult to quantify).
Iconsider the recent inclusion of Trichoplusia oxygramma in Agrapha by Franclemont & Todd (1983) to be
incorrect as this species has none of the autapomorphies of Agrapha just enumerated. Similarly, I reject
their synonymy of Ctenoplusia and Acanthoplusia with Agrapha. No evidence was found in this study to
suggest that these three genera form a monophyletic unit.

Agrapha is the only argyrogrammatine genus wholly endemic to the New World, being found from Costa
Rica to Peru. A. calceolaris and a possibly undescribed species (not sp. A, may or may not be sp. B) occur
on Jamaica. Nothing is presently known about the biology of any of the species but they appear to be
widespread through a range of altitudes. A. meretricia is recorded from 320 ft (100 m) in Colombia to
6500+ ft (2000+ m) in Peru. Most specimens in the BMNH collection have been taken above 2500 ft
(750 m).

Clade 15 [92(3), 144(0)]. The five genera of this clade are united solely on the basis of their possessing an
attenuate and dorsally recurved sub-basal lobe on the lateral arms of St8 (92). The degree of development
of the clavus (144) has already been shown to be a worthless character (see Character analysis).

Clade 16: Ctenoplusia s.1. [4(2), 60(1), 77(1), 119(1), 133(1), 797(1)]. Given that blade-like saccular setae
(133) and a pleurite partially fused with the VAVB (119) are interpreted as convergently derived in this
clade and in Plusiotricha, only two good synapomorphies were found to unite Ctenoplusia and Acanthoplu-
sia: the tendency for the pink forewing terminal line to be concentrated at the apex of vein M; (4) and the
totally desclerotized, membranous antrum (197). The former is by no means universal in Ctenoplusia but
does occur (or its more derived state 3) in C. limbirena, C. amydra Dufay, C. accentifera and the C.
furcifera (Walker) species-complex. This corroborates the hypothesis that Acanthoplusia is the sister-
group of a section of Ctenoplusia and supports Ichinosé’s (1973) synonymy. The occurrence of the
desclerotized antrum in other species of Crenoplusia and Acanthoplusia needs to be confirmed.

_ Acanthoplusia [83(3), 84(0), 108(0), 214(0)]. Acanthoplusia was erected by Dufay (1970b) to accommo-
date five species: A. tarassota (Fig. 53) (NE. Himalaya, Java, Bali), A. ichinosei (Dufay) (Japan), A.
— agnata (China, Japan), A. latistigma (Prout) (Ceram) and A. eugrapha (Hampson) (New Guinea).
Subsequently, other species were included (Dufay, 1970b; 1974; 1982): A. adiaphora Dufay (Taiwan), A.
herbuloti Dufay (Luzon, Philippines), A. sigillata Dufay (Luzon, Philippines) and A. vermiculata (Sumat-
ta). These species differed from Ctenoplusia in the number of cornuti, the shape of the male tergites of
segments A5 and A6, and the presence of a pad of inflated scales on the outside of the valve. The first two
characters were found to be too variable within the Plusiinae as a whole to be satisfactorily coded in the
present study but may have value in a species-level investigation. The scale pad is, however, a good and
easily recognizable apomorphy for Acanthoplusia, which is doubtless a monophyletic taxon, even if not
finally deserving full generic status.

Three species are known from immature stages: A. tarassota, A. ichinosei and A. agnata. All three
appear to be somewhat polyphagous (Ichinosé, 1962b; Miyata, 1983). The larvae of A. agnata are unusual
in the Argyrogrammatini in completely lacking prolegs on segments A3 and A4, instead of bearing
vestigial, peg-like structures (Ichinosé, 1962b; Eichlin & Cunningham, 1978). All species of Acanthoplu-

sia, except A. agnata and A. latistigma, have a similar, very finely reticulate, brown forewing pattern.

Ctenoplusia (s.str.) [4(3), 60(2), 78(2), 86(0)]. As mentioned above (Character analysis, character 60),
Ctenoplusia (s.str.), as currently constructed, is probably a polyphyletic assemblage of those species with
blade-like saccular setae (and some that lack them) that are not members of Acanthoplusia or Plusiotricha.
The last genus shares a similar form of basitarsal comb not found in Acanthoplusia (see character 60,
Character analysis), while the conspicuous pink forewing spot appears to be a potential synapomorphy for
C. limbirena (Fig. 54) and several other species (see character 4, Character analysis). The distribution of
dorso-lateral tufts that are enclosed in a pleural fold (78) needs to be checked across a wider range of
species.

Of the 50 or so species of Ctenoplusia (s.str.), the life histories of only two appear to have been
elucidated. C. limbirena, like many plusiines, is polyphagous, feeding on Althea (Malvaceae), Becium,

148 IAN J. KITCHING

Salvia (both Lamiaceae), Primula (Primulaceae), Verbascum (Scrophulariaceae), Solanum and Nicotiana
(both Solanaceae) (Pinhey, 1975). In contrast, C. albostriata (which is possibly not congeneric with C.
limbirena) is apparently restricted to members of the Asteraceae (Gardner, 1947; Robinson, 1975;
Miyata, 1983; Holloway, 1985), although there are records for Calystegia (Convolvulaceae) (Miyata, 1983)
and Symphytum (Boraginaceae) (Holloway, 1985). Further support for a close relationship between C.
albostriata and Trichoplusia vittata and T. oxygramma (see 60, Character analysis) is provided by Pinhey’s
(1975) record of Commidendron (Asteraceae) as the foodplant of T. vittata (as transfixa) and Crumb’s
(1956) records of Aster, Solidago, Erigeron (Asteraceae) and Nicotiana as foodplants of T. oxygramma. In
C. albostriata, a form is known (f. innotata) in which the modified Y-mark is missing. The genetics of this
and the typical form (f. a/bostriata) was investigated by Ichinosé et al. (1967). The two forms were found to
obey Mendel’s laws of heredity, with f. albostriata being totally dominant to f. innotata.

Clade 17 [61(1), 83(4), 87(0), 138(0)]. Plusiopalpa, Stigmoplusia and Argyrogramma are united as clade 17
on the basis of character 61(1), a suppressed inner SV row of setae on the male foretarsi. This state is,
however, convergently derived in Plusiotricha.

Plusiopalpa [12(1), 47(1), 84(0), 86(0), 91(1), 128(1), 201(1), 213(2)]. Plusiopalpa is a compact genus of
five species that have an unmistakable general appearance. In addition to large size and a somewhat
distinctive colour pattern, their most obvious feature is that which has given the genus its name, the very
long labial palps. This is due to the extreme dorso-apical elongation of the third segment (41, see Character
analysis). In addition there are numerous generic autapomorphies of the genitalia of both sexes including a
scimitar-shaped valve, vinculum arms with a marked dorsal keel and medial constriction, and a distinctive
corpus bursae in which the ductus seminalis arises from a small apex produced laterally about halfway
down the corpus bursae, near the point of entry of the ductus bursae.

Plusiopalpa occurs throughout the Old World tropics. P. thaumasia Dufay and P. hildebrandti
(Saalmiiller) are endemic to Madagascar, while P. shisa appears to be restricted to Taiwan. P. adrasta
ranges throughout the Oriental tropics to New Guinea, and P. dichora (Fig. 58) occurs in Africa.
Hampson’s (1913) synonymy of P. adrasta and P. dichora, followed by Gaede (1940) and Pinhey (1975), is
incorrect (Dufay, 1970a), while the separation of P. shisa from P. adrasta ought to be confirmed by
genitalic dissection. At present, the only diagnostic characters are minor wing pattern elements. Hampson
(1913) also included a sixth species in Plusiopalpa, P. camptogamma (Hampson). Dufay (1970a) disagreed
with this placement but did not suggest a better one. Examination of the genitalia of the female holotype
showed that it does not possess the characteristic form of the corpus bursae referred to above that is found
in other plusiopalpas, but has a simple corpus bursae instead. Thus P. camptogamma should be removed
from Plusiopalpa. Dufay (unpublished combination in BMNH collection) has proposed that the species is
properly placed in Crenoplusia, and the possession by the male of a quadrisetose basitarsal comb (see
Character analysis) supports this move. However, examination of the male genitalia is required to confirm
this placement as correct. Only P. adrasta has ever been reared, from Mikania, an asteraceous plant
introduced into Peninsular Malaysia (Holloway, 1985).

Clade 18 [60(3), 61(2), 79(3), 85(1), 90(1), 92(2), 93(3), 94(3), 130(0), 162(1)]. Although outwardly
different in general appearance, Argyrogramma (s.str.) and Stigmoplusia share a remarkable number of
synapomorphic features, all of which are probably secondary sexual characteristics of the males. These are:
a basitarsal comb composed of multiple rows of setae (60); a further suppression of the foretarsal SV spines
(61); dorso-lateral hair tufts with discrete edges on segment A6 only (79, but see Agrapha and Stigmoplu-
sia); differentiated sternal hair pencil insertions (85); an anterior flange on the lateral arm of St8 at the
point where they curve dorsally over the hair pencils (90); lateral arms of St8 with the sub-basal lobe
modified into a blunt projection (92); dorsal arms of St8 that run almost transversely (93); and apical scales
on the dorsal arms of St8 modified into a semicircular, projecting plate (94).

Argyrogramma (s.str.) [61(3), 78(3), 88(1), 98(1), 105(1), 108(0), 121(2)]. The members of the genus
Argyrogramma (s.str.) are most easily recognized by a feature of the male genitalia — the vinculum arms
bear broad dorsal and ventral flaps that are folded horizontally and partially fused together to give a stocky,
rectangular saccus that is markedly truncate anteriorly (e.g. A. signata; Dufay, 1970a: fig. 96). In this
study, two autapomorphies were found for A. verruca, characters 61 (completely suppressed male
foretarsal SV spines; short, stout V spines) and 78 (dorso-lateral hair tufts on A7 enclosed in a membranous
fold). Two further character states (88, 98) were shared only with Agrapha (q.v.) and one only with
Abrostola (121). Their occurrence in the other members of the genus needs confirmation.

Argyrogramma (s.str.) currently contains 12 species. However, when first described (Chou & Lu, 1978;
1979a), five were placed in Argyrogramma in the broad sense (i.e. sensu Ichinosé, 1973). These species are

SYSTEMATICS OF THE PLUSIINAE 149

more likely to prove to be ctenoplusias or thysanoplusias and I will not consider them further. A.
circumscripta does not agree with the type species, A. verruca, either. However, Kostrowicki (1961), who
transferred circumscripta to Argyrogramma, had a very peculiar concept for the genus and this species also
probably belongs elsewhere. Similarly, A. aenofusa was transferred to Argyrogramma (s.1.) by Chou & Lu
(1978). However, this species does share a potential apomorphy with A. verruca, namely a basitarsal comb
composed of multiple rows of setae (60:3, see Character analysis). Thus, for the moment, I will leave A.
aenofusa in Argyrogramma (s.str.), despite numerous other differences (e.g. highly modified, spatulate
saccular setae).

The core of Argyrogramma (s.str.) can thus be considered to consist of five species. Of these, A.
hainanensis Chou & Lu is almost certainly synonymous with A. signata, a widespread species in the Old
World tropics, found from the Canary Islands, through to Tonga and the Cook Islands (Dufay, 1970a;
Holloway, 1985). A. subaerea Dufay is recorded from widely scattered localities in tropical Africa from the
Ivory Coast to Uganda and SW. Angola (Dufay, 1972), while A. verruca (Fig. 59) and A. basigera occur
throughout North America and Mexico, with the former extending down to Peru and Argentina (Eichlin &
Cunningham, 1978), even reaching the Galapagos Islands (Hayes, 1975).

The larvae of both A. verruca and A. basigera have been described by Eichlin & Cunningham (1978).
The former is polyphagous while the larval hostplant of the latter has not been recorded. Although the
immature stages of A. signata have not been described, Holloway (1985) stated that the species has been
reared on Eucalyptus in New Guinea. Given that most plusiines feed on herbaceous plants, this record
requires confirmation as a natural hostplant.

Stigmoplusia [11(1), 79(2), 83(3), 97(1), 111(1), 112(2), 144(1), 214(0)]. Three synapomorphies were
found for Stigmoplusia chalcoides in the present data set: discrete edges to the dorso-lateral hair tufts on
segment AS only (79) and a vinculum produced as a long spine (111; similar to the condition found in clade
8, q.v.), bearing two sharp, subdorsal points (112). There are also several autapomorphies not contained in
the data set, including a highly differentiated sternal hair pencil comprising several types of scales (Fig.
270); broad, subventral flaps on the uncus (Figs 280, 281); uncus apically rounded-triangular in cross-
section; and strong, asymmetrical digitations on the ventral edge of the valve, distal to the apex of the
sacculus, which terminates in a stout, curved spine (Fig. 353).

Seven species are currently placed in Stigmoplusia, all described by Dufay (1970a; 1972; 1975): acalypta,
allocota, antsalova, chalcoides (Fig. 60), epistilba, megista and paraplesia. All except S. acalypta and S.
megista possess uncal flaps. In addition to S. chalcoides, valve digitations are present in S. allocota, S.
acalypta and S. megista, but occur along the entire ventral edge of the valve, each digitation bearing a

- single, short, apical seta. S. epistilba and S. paraplesia have simple valves, while those of S. antsalova are

very peculiar (see Dufay, 1970a: fig. 109); the apex is enlarged with a basally-oriented, rounded lobe that
bears an apical membranous sac. As far as can be ascertained, all species have vinculum arms with
subdorsal, sharp points. Stigmoplusia thus presents quite a diverse array of male genitalic structure. The
females of only three species are known (S. chalcoides, S. allocota, S. acalypta) and all agree in the apex of
the corpus bursa being ornamented internally with a conspicuous, sclerotized, reticulate pattern of ridges.

Nothing is known about the biology of any species of Stigmoplusia. The genus is found in the tropical
forest belt of Africa from Liberia to Uganda and Zimbabwe. S. antsalova, with the unusual male genitalia,
is endemic to west Madagascar.

Clade 19: Plusiini [117(1), 121(1), 127(1)]. The members of the primarily Holarctic tribe Plusiini are united
by the synapomorphic possession of a broad flange on the pleurite (117). This structure is, however,
subsequently lost in the highly autapomorphic Antoculeora. A juxta with an apical notch (or subsequent
derivation) (127) is also characteristic of this tribe, although it is lost in Allagrapha and convergently
derived in Omorphina and Chrysodeixis.

The larvae of the Plusiini are characterized by the total loss of the prolegs on segments A3 and A4 (but
these are also convergently lost in Acanthoplusia agnata, see above). There also appears to be a greater
degree of larval oligophagy in the Plusiini, but this may be due solely to a lack of information for many
(particularly non-pest) members of the Argyrogrammatini.

Clade 20: Euchalciina [39(1), 83(0), 101(1), 178(2)|. The genera of the plusiine subtribe Euchalciina are
characterized by two good synapomorphies: the pleural pouch is small and concealed within the tympanal
hood (178) and all have a distinctive outline to the tegumen in lateral view (101). In addition, the sternal
hair pencils are invariably reduced to a mere sparse tuft of scales (83), although this reversal to the
plesiomorphic condition has also taken place in other genera (Pseudoplusia, Eutheiaplusia, Syngrapha).
| Finally, labial palps in which the ventral scales form a neat ridge (39) are also characteristic of the

|

150 IAN J. KITCHING

Euchalciina, but the condition appears to have been lost in Euchalcia and Adeva (clade 25). Most of the
euchalciines feed as larvae on Ranunculaceae, which may prove to be a further synapomorphy for the
subtribe.

Eosphoropteryx [175(1), 213(1)]. Eosphoropteryx is a monobasic genus, comprising E. thyatiroides (Fig.
61). It has a number of genitalic autapomorphies, including a keel-shaped valve with a produced apex; an
unsclerotized, sub-basal auricular cornutus on the vesica (175; similar cornuti occur in Macdunnoughia and
Loboplusia, but are heavily sclerotized in these genera); a flattened, laterally curved, heavily sclerotized
antrum (Fig. 452); and granular ridges on the apex of the corpus bursae (213).

E. thyatiroides is a pinkish-grey plusiine, found across the northern United States and southern Canada,
from Nova Scotia and Virginia to British Columbia and Oregon (Eichlin & Cunningham, 1978). The larva
is reported to feed on Thalictrum (Ranunculaceae) (Forbes, 1954) but no other details of the species’
biology are known.

Clade 21 [109(1), 147(1), 196(1)]. The remaining genera of the Euchalciina, excluding Eosphoropteryx, are
united primarily by the synapomorphic shape of the valve (147). The simple, oval valve occurs only in clade
21, with the exception of a convergent occurrence in Omorphina (q.v.). As noted above (Character
analysis), the valve of Euchalcia variabilis is more quadrate than oval, but the majority of euchalcias show
the oval valve form. An S-shaped vinculum apex in lateral view (109) is weakly expressed in this clade and
also occurs in clade 39 (Plusiina part) and in Cucullia. A broad, shallow, open, bowl-like antrum (196) is
also characteristic of all members of clade 21 but is highly homoplasious, occurring additionally in Agrapha
and clade 8 within the Plusiinae and in Diloba, Cucullia and Chalcopasta in the outgroups.

Polychrysia[5(1), 41(1), 50(1), 65(2), 165(1)]. One unique and unreversed apomorphy for Polychrysia was
found in the present data set; the presence of a few, fine SD and D setae on the tarsi (65). It is unknown if
this character state is present in the other members of the genus. Polychrysia also possesses scattered black
forewing scales (5), an elongate labial palp segment 3 (41) and erect tegulae (50), all of which occur
(homoplasiously) only in the Euchalciina (with the exception of the first, which is also found elsewhere in
the Plusiini). The subapical cornutus borne on a lateral diverticulum (165) may represent another
apomorphy for the genus. The superficially similar condition seen in Diachrysia is almost certainly not
homologous.

Polychrysia currently contains six species. The most widespread is P. moneta (Fig. 62), known in Britain
as the Golden Plusia. It occurs across Canada from Ontario to Alberta, and north to Alaska, but has yet to
be recorded from British Columbia or east of Lake Huron (Eichlin & Cunningham, 1978). Widely
distributed in Europe, P. moneta is replaced east of the Urals by P. esmeralda (Oberthiir). This species
differs from P. moneta in the silver-grey ground colour to the forewing, rather than a golden-yellow, and
has only recently been reinstated as a good species (Kostrowicki, 1961; Chou & Lu, 1979a). However,
there is some confusion (compare the figures of the male genitalia in the above two publications) and the
separation of P. esmeralda from P. moneta is still open to question.

P. sica, P. aurata (Staudinger) and P. splendida (Butler) are all eastern Palaearctic (China, Japan,
Siberia) (Kostrowicki, 1961), while P. morigera (Edwards) is restricted to the foothills of the Rocky
Mountains of the U.S.A., both the eastern and western slopes. ‘Polychrysia’ aureus (Bang-Haas)
(Kostrowicki, 1961) is not a plusiine but is correctly assigned to the amphipyrine genus Chrysonicara
Draudt, near Nicara Moore (Draudt, 1937).

The life history of P. moneta in Europe is well known (Bretherton er al., 1983). The larvae are most
frequently recorded feeding on members of the Ranunculaceae; e.g. Aconitum, Delphinium (South, 1961;
Skinner, 1985) and Trollius (Beck, 1960; Kostrowicki, 1961). However, the last author also gives Cucumis
(Cucurbitaceae) and Helianthus (Asteraceae), while South (1961) adds Arctium (Asteraceae). P. splendi-
da appears to be restricted to Aconitum (Miyata, 1983), while P. morigera is recorded only from
Delphinium sp. (Eichlin & Cunningham, 1978). Pupation is in a characteristic strong, oval cocoon of
yellow silk spun between leaves of the foodplant (Bretherton er al. , 1983).

Clade 22 [3(0), 154(1), 183(1)]. The members of clade 22 are primarily characterized by short anal papillae
(183). However, as noted above (Character analysis), Euchalcia variabilis exhibits the plesiomorphic
condition of approximately square papillae and this was interpreted as an early stage in a transform-
ation series culminating in the blade-like papillae of species such as E. augusta. Transtillae that are fused
above the aedeagus into a broad, blunt point (154) also characterize clade 22, with the exception of the
included clade 26. Further investigation may show this character state to be a synapomorphy of clades 23
and 25.

SYSTEMATICS OF THE PLUSIINAE 151

Clade 23 [42(1), 132(1), 163(1)]. Two synapomorphies unite Panchrysia and Chrysanympha into a
monophyletic group: fused saccular flanges modified into a rounded lobe between the valves (132) and an
apically swollen subapical cornutus (163). However, the latter appears to be present in Panchrysia only in
P. deaurata, P. ornata (Bremer) (Kostrowicki, 1961; Chou & Lu, 1978) and P. mishanensis Chou & Lu
(Chou & Lu, 1979a), while the former may be of more general occurrence, being distinguishable in figures
of the genitalia of P. deaurata, P. ornata and P. dives but not P. v-argenteum (Esper) (Kostrowicki, 1961)
or P. mishanensis (Chou & Lu, 1979a). The sister-group relationship of Panchrysia and Chrysanympha
therefore remains equivocal.

Chrysanympha [5(1), 41(1), 107(1)]. Chrysanympha contains a single species, C. formosa (Fig. 63), of
unique general appearance within the Plusiinae. The forewing has a strong, brown, evenly-curved
subterminal line and the pinkish grey colour of the Y-mark is extended costally and basally to give a
superficially similar pattern at the wing base to Abrostola (into which genus C. formosa was incorrectly
placed by Hampson, 1913). Chrysanympha also bears a dorsal sclerotized bar on the anal tube (107),
similar to that seen in Lophoplusia and the Abrostolini, and is another euchalciine to display scattered
black scales on the forewing (5).

C. formosa appears to be a scarce species confined to the north-eastern U.S.A. and south-eastern
Canada (Eichlin & Cunningham, 1978). Its life history has been briefly described by Kearfott (1904) and
the larval foodplants are given as Gaylussacia and Vaccinium (Ericaceae) (Tietz, 1972), unusual for the
Euchalciina.

Panchrysia[19(1),65(3), 209(1)]. Panchrysia deaurata is the only euchalciine to my knowledge in which the
ductus seminalis arises from the fundus of the corpus bursae (209). This may prove to be an autapomorphy
for the species, as the ductus seminalis has the more usual origin from the apex of the corpus bursae in P.
ornata (Chou & Lu, 1978: fig. 23). Like Polychrysia, Panchrysia deaurata has SD and D tarsal setae (65)
but in the present genus, they are numerous and long, much as in the distantly related Lophoplusia. The
presence of this character state in the other members of the genus needs confirmation.

Five species currently comprise Panchrysia. The type species, P. deaurata (Fig. 64), is found in steppe
and semi-arid areas of Europe and Central Asia from Spain to the Altai. It is a golden-yellow moth,
superficially(?) similar to Polychrysia aurata, and thus unlike the other members of its genus in general
appearance. The other four species all bear a silver Y-mark and other metallic spots on the forewings. P.
v-argenteum (Fig. 65) is a pinkish brown moth restricted to the European Alps and the high mountains of
south-east Europe (e.g. Yugoslavia, Greece), while P. ornata is very similar but pale grey-brown. The
latter species occurs in the eastern Palaearctic steppes (Kostrowicki, 1961), and thus stands in relation to P.
v-argenteum, in terms of both distribution and colour, much as Polychrysia esmeralda stands in relation to
Polychrysia moneta. Panchrysia mishanensis is very similar to P. ornata (Chou & Lu, 1979a) and
presumably originates from Mi-Shan in far north-eastern China, but this is not certain. The fifth species, P.
dives (Fig. 66), is similar to P. v-argenteum but the forewing is chocolate-brown, has more extensive silver
markings, and the hindwing is yellow with a brown margin. It is largely sympatric with P. ornata, but
extends further east into southern Kamtchatka and is absent from the Korean peninsula and most of China
(Kostrowicki, 1961).

Little is known of the biology of Panchrysia. The larval foodplants of P. deaurata are recorded as
numerous species of Thalictrum (Ranunculaceae), while P. v-argenteum feeds on Jsopyrum (Ranuncu-
laceae), as well as Thalictrum foetida (Kostrowicki, 1961).

Clade 24 [6(0), 144(1), 768(7), 205(1)]. The five genera comprising clade 24 are united by the synapomor-
phic possession of a vesica that is internally granular basally (168).

Clade 25 [4(4), 39(0), 58(0), 164(1)]. Euchalcia and Adeva are considered sister-taxa on the basis of two
synapomorphies. In both genera, the terminal line on the forewing is white not pink (4) and the subapical
cornutus on the vesica arises from a circular, concave base-plate (164). These two character states also
occur in all other Euchalcia species represented in the BMNH collection or figured by Dufay (1968), with
the following exceptions. The white terminal line is absent in E. inconspicua (Graeser) (Fig. 69), E.
cashmirensis Moore and E. dorsiflava, and is reduced in several other species. The circular base-plate of
the subapical cornutus is absent only in E. inconspicua, where the subapical cornutus is replaced by a series
of cornuti, and E. tancrei (Staudinger). Dufay (1968) considered Euchalcia to be a very homogeneous
genus, in terms of general facies and male and female genitalia, providing E. inconspicua, E. tancrei and
‘Plusia’ hampsoni Leech (Fig. 68) were excluded (all were first transferred to Euchalcia by Kostrowicki,
1961, the last as its synonym, ‘Phytometra’ /Jenzeni Draudt [Dufay, 1965]). Euchalcia inconspicua has

152 IAN J. KITCHING

recently been placed in its own genus, Pseudochalcia, by Kljutshko (1984). I have examined the male
genitalia of hampsoni (BM noctuid slide no. 11521) and agree with Dufay (1968) that this species does not
belong in Euchalcia. Iam unable at present, however, to suggest where ‘P.’ hampsoni ought to be placed. I
superficially examined specimens of Euchalcia tancrei, together with the drawing of the male genitalia in
Kostrowicki (1961), and I concur with Dufay that this species, also, does not belong in Euchalcia. Although
it is unsatisfactory, until such time as a more comprehensive study is performed, both hampsoni and tancrei
should be placed in limbo (i.e. ‘Plusia’ s.1.). This action, together with the removal of inconspicua to
Pseudochalcia (see also Consensus classification), almost certainly leaves Euchalcia (+ Adeva)
monophyletic.

Euchalcia [183(0), 201(0)]. The genus Euchalcia comprises around 30 species. They are found almost
exclusively in the steppe and semi-arid regions of the Palaearctic, the exceptions being E. cashmirensis, E.
serraticornis Dufay and E. xanthoides Dufay, which are subtropical-Himalayan species (Dufay, 1968).
Euchalcia gerda and E. serraticornis have serrate male antennae (the females are unknown for both
species), and are thus somewhat unusual (Dufay, 1968).

The biology of only seven species is known (Dufay, 1968). E. variabilis (Fig. 67) and E. bellieri (Kirby)
are reported to feed as larvae on the usual ranunculaceous foodplants of the Euchalciina, Aconitum and
Thalictrum. However, one group of euchalcias has undergone a foodplant switch to members of the
Boraginaceae: Alkana (E. chlorocharis (Dufay)), Cynoglossum (E. cuprea (Esper) and E. italica (Staudin-
ger)), Lycopsis (E. consona (F.)), Pulmonaria (E. consona and E. cuprea), Rindera (E. siderifera
(Eversmann)), Nonea (as Nonnea; E. consona) and Symphytum (E. cuprea) (Beck, 1960; Kostrowicki,
1961; Dufay, 1968). In addition, E. cuprea is recorded from Pterotheca (Asteraceae) and, most unusually,
Arum (Araceae) (Kostrowicki, 1961).

A comprehensive revision of the genus has been provided by Dufay (1968), including detailed accounts
of the morphology, distribution and affinities of 29 species of Euchalcia and separate keys to species using
general habitus, male and female genitalia.

Adeva[5(1), 59(0)]. Adeva is yet another euchalciine to have scattered black scales on the forewing, and in
this character state, the sole included species, A. albavitta, differs from all species of Euchalcia that I have
examined.

Adeva was erected by McDunnough (1944) to accommodate A. albavitta (Fig. 71). The genus was
considered synonymous with Euchalcia by Kostrowicki (1961), a decision endorsed by Dufay (1968). A.
albavitta is restricted to the south-western U.S.A., where it occurs in the states of Oregon, California,
Arizona and Nevada (Eichlin & Cunningham, 1978). There are two colour forms, a pale brownish form
found in the coastal regions of California, and a smoky grey desert form, originally described in the genus
Behrensia Grote (McDunnough, 1944). Nothing is known about the biology of A. albavitta. However,
given the dichtomy in Euchalcia of larval foodplants between the Ranunculaceae and the Boraginaceae,
knowledge of that of Adeva might prove useful in resolving the position of this species with regard to the 30
or so euchalcias.

Clade 26 [144(2), 154(0), 771 (1), 186(1), 194(2)]. Plusidia, Lamprotes and Pseudeva form a monophyletic
group on the basis of three synapomorphies: the presence of an isolated, dorsal patch of cornuti on the
vesica (171); short, thick membrane preventing easy protraction of the anal papillae (186); and the
presence of several setae along the inner, ventral edge of the antrum (194). This last character state is
further transformed in Lamprotes (q.v.) and a superficially similar but non-homologous condition occurs in
Diloba.

Plusidia [58(0)]. Plusidia currently contains three species. However, P. imperatrix and P. chinghaiensis
Chou & Lu show considerable similarities in wing pattern to Polychrysia splendida and examination of the
genitalia may show that they are properly placed in Polychrysia.

This action would render Plusidia monotypic, containing only P. cheiranthi (Fig. 34). At one time, the
eastern Palaearctic populations were considered to be a separate species, P. separanda Warren (Inoue &
Sugi, 1958-61). More recently, it has been relegated to subspecies of P. cheiranthi (Kostrowicki, 1961) and
finally considered to be fully synonymous (Sugi, 1982).

Western populations of P. cheiranthi are unmistakable in appearance. The forewings are pinkish grey,
with a chocolate-brown, comma-shaped apical spot and a similar brown spot at the base. However, eastern
examples have a much more ‘washed-out’ appearance with the apical spot barely distinguishable and the
basal spot often somewhat faded also. The larva was briefly described by Kirby (1889), who gave the
foodplants as Thalictrum and Aquilegia (Ranunculaceae).

SYSTEMATICS OF THE PLUSIINAE 158

Clade 27 [6(1), 16(1), 19(1), 41(1), 42(1), 205(0)]. There are no good synapomorphies to support the
sister-group relationship between Lamprotes and Pseudeva, which must thus be considered tentative at
best. Nevertheless, this relationship has been proposed previously by Kostrowicki (1961), who stated that
Lamprotes was ‘in facies and genitalia much like the species of the Nearctic genus Pseudeva’ (see also
Pseudeva below).

Lamprotes [3(1), 144(1), 162(0), 191(1), 194(1), 201(1)]. Lamprotes can be separated from the other two
genera of clade 26 by the restriction of the ventral antrum setae to a subventral or lateral position (194). In
addition, most Lamprotes specimens bear a small, golden, C-shaped mark below the forewing cell (3). This
marking was coded in the data as homologous with the Y-mark of other plusiines. However, the C-mark of
Lamprotes occupies a slightly more basal position (entirely below vein CuA) rather than straddling it) and
this suggests that it represents a novel pattern element that is autapomorphic for Lamprotes.

Lamprotes has a very similar disjunct distribution to Plusidia, with western and eastern Palaearctic
populations. However, in contrast to that genus, the eastern population of Lamprotes, L. mikadina Butler,
has long been treated as a good species (Hampson, 1913; Inoue & Sugi, 1958-61; Chou & Lu, 1979a; Sugi,
1982). Occasionally, L. mikadina has been treated as a subspecies of L. c-aureum (Kostrowicki, 1961) (Fig.

35), or even fully synonymized (implicitly in Forster & Wohlfahrt, 1971).

The larva of L. c-aureum has been briefly described by Kirby (1889) and that of L. mikadina in more
detail by Ichinosé (19626). Both species are reported to feed on Thalictrum and Aquilegia.

The generic name Lamprotes is unusual insofar as the identity of the author is unknown, only his or her
initials accompanying the first reference to the name (R.L., 1817). According to W. H. T. Tams (Nye,
1970; unpublished notes), the identity of R.L. was investigated by C. D. Sherborn. However, he reached
no firm conclusion. C. Boursin has attributed authorship to Reichenbach of Leipzig, but it could also be
Rudolph Leuckhart (who published under his own name 1836-1859) or Rudolph Ludwig, suggested by E.
L. Todd (Nye, 1970; unpublished notes). Nye suggested that continuing to use R.L. was the most prudent
course.

Pseudeva [5(1), 50(1)]. Pseudeva contains only two species, P. purpurigera (Fig. 36) and P. palligera. Both
species, like Polychrysia, have erect tegulae (S50). However, this character state is difficult to score in any
specimen that is little less than perfect and may yet be shown to occur in other euchalciine genera.

Both species of Pseudeva are endemic to the Holarctic region. P. palligera occurs only in the
south-western U.S.A., while P. purpurigera is widespread in southern Canada and the northern two-thirds
of the U.S.A., east of the Rocky Mountains, but is noticeably absent from the central band of states from
North Dakota to Texas.

The larva of P. purpurigera has been described by Dyar (1896), who gave Thalictrum as the foodplant.
He also stated that the larva was ‘green and white [and] curiously hunched up’. This description is very
similar to the coloration and posture of the larva of Lamprotes mikadina (Ichinosé, 1962b: pl. 18, fig. 4;
Mutuura eral. , 1965: pl. 37, fig. 113) and these features may represent good synapomorphies for Lamprotes
and Pseudeva. Further examination of the larvae of Pseudeva is thus important in confirming the
sister-group relationship of these two genera.

Clade 28 [116(1)]. The subtribes Plusiina and Autoplusiina are interpreted as sister-taxa solely on the basis
of a strongly curved pleurite (116). As noted earlier, such a structure also occurs in Anadevidia and thus the
monophyly of clade 28 can only be considered to be weakly supported. Eichlin & Cunningham (1978)
treated the Plusiina and (paraphyletic) Euchalciina as sister-taxa on the basis of a shared fusion of the
pinacula of setae SV1 and SV2 on segment A2 of the larva. However, this character state also occurs in
several non-Nearctic autoplusiine genera (Macdunnoughia, Sclerogenia and Erythroplusia), as well as in
the argyrogrammatines Thysanoplusia and Anadevidia (Ichinosé, 1962b), and is thus of dubious value.

Clade 29: Autoplusiina [18(0), 213(2)]. The members of the Autoplusiina are only weakly united as a
monophyletic group on the basis of sclerotized ridges at the entry point of the ductus bursae into the corpus
bursae and on the apex of the corpus bursae (213). However, these ridges must have been secondarily lost
in Erythroplusia. In addition, such ridges occur in Omorphina and Plusiopalpa within the Plusiinae, and in
Diloba and Chalcopasta of the outgroups. Further study is required into this character and, given the
reservations regarding the homologies of other character states important in the classification of this
subtribe (e.g. see Characters, 142: 1 & 2), the Autoplusiina may subsequently be shown to be paraphyletic
Or worse.

Rachiplusia [14(2), 59(0), 63(1), 64(2), 110(1), 114(1), 139(1), 148(0), 154(1), 191(1), 198(3), 206(1),

154 IAN J. KITCHING

216(1)]. Rachiplusia is perhaps the most problematical autoplusiine genus, in that it shares possible
synapomorphies with a number of subgroups of the subtribe. Rachiplusia and ‘Syngrapha’ gammoides are
the only autoplusiines that are sexually dimorphic (216), while a bisetose female frenulum (14) and spined
hind tibiae (64) further corroborate a relationship between Autoplusia and Rachiplusia. Rachiplusia also
shares character states with larger clades that include Aufoplusia: increased sclerotization of the LPV and
T8 ventrally (198: clade 35); lack of a harpe (140: clade 34); and a protected ostium bursae (191: clade 36).
This last character is also shared, however, with the members of clade 32, as is a clavus base that is obscured
by a saccular overfold (139). Finally, Rachiplusia and Macdunnoughia s.1. (clade 33) both have a medially
expanded section of the ductus bursae with dorsally recurved edges (206). It is thus unlikely that
Rachiplusia will prove not to be related to some or all of the autoplusiines of clade 30. However, the genus
lacks many structures found in this clade, most notably the apomorphic states of characters 95, 141, 189 and
193 (see below). A more extensive survey of the Autoplusiina (and maybe the Plusiina, see Anagrapha and
Syngrapha below) is necessary to resolve these ambiguities.

Rachiplusia comprises three species. Two, R. nu (Fig. 72) and R. ou. are very similar, both in general
appearance and in genitalia. R. ou occurs from the U.S.A. south to Venezuela (including Jamaica and the
Bermuda Islands), while R. nu is found in Bolivia, Chile, Brazil and Argentina. The immature stages of R.
nu have been described in detail by Angulo & Weigert (1974), and the larvae of R. ou by Crumb (1956).
Both species are highly polyphagous (Angulo & Weigert, 1975; Eichlin & Cunningham, 1978) but only R.
nu appears to be of economic importance as an agricultural pest (Angulo & Weigert, 1975).

The third species, R. virgula, has long been considered a member of the genus Caloplusia (e.g.
Hampson, 1913; Kostrowicki, 1961), because of its reduced eyes, yellow hind wings and fully spined tibiae.
Following examination of the genitalia, however, Angulo (1978) transferred virgula to Rachiplusia. There
can be little doubt that this reflects the relationships of the species more precisely. Both R. virgula and R. |
nu (the second species figured by Angulo, 1978) have a ventrally bulged valve (Fig. 360); long, thin clavi
(Fig. 360); a broad, truncate saccus (Fig. 302); and a similar arrangement of cornuti on the vesica. The
immature stages of R. virgula have been described by Izquierdo (1895), who gave the larval foodplant as
Madia sativa (Asteraceae). R. virgula is endemic to the Chilean Andes.

It does not appear to have been noted before that two of the species of Rachiplusia are sexually
dimorphic. The males are more brightly and contrastingly coloured than the females, which are dull in
comparison. In addition, the males have a predominantly grey forewing ground colour, while the females
are brown. A similar but more striking dimorphism occurs in ‘Syngrapha’ gammoides (Figs 82, 83). In
contrast, R. virgula appears to be monomorphic, although I have not seen a sufficiently long series of this
species to be certain.

Clade 30 [83(2), 95(1), 141(1), 162(0), 189(1), 193()]. The remaining members of the Autoplusiina,
excluding Rachiplusia, are united by a single unique and unreversed synapomorphy, the presence of short,
upright scales on the antero-ventral margin of the antrum (193). In addition, a broad, triangular saccular
spine (141) occurs in many members of the subtribe, although some extreme modifications may eventually
prove to be incorrectly homologized (e.g. the quadrate lobe of Erythroplusia and the lance-like projection
of Antoculeora). Fine, hair-like scales ventrally between the lateral arms of St8 (95) occur in all the member
genera of clade 30 except Sclerogenia, while a T8 that is cut back ventral to the anterior apophysis (189) is
present in all but Antoculeora, where the extreme autapomorphic condition of the male genitalia make it
difficult to score the character. However, the apomorphic state of character 189 also occurs in Syngrapha
(s.1.) and Anagrapha (see below). Again, a more extensive investigation is necessary, particularly of the
male genitalia, in order to confirm several of the homologies hypothesised in this study (especially those of |
characters 141 and 142).

Erythroplusia [5(1), 6(1), 18(1), 104(1), 108(1), 145(1), 155(1), 158(2), 209(1), 213(0)]. Like Rachiplusia,
the relationships of Erythroplusia are somewhat obscure. It also possesses character states that associate it |
with several different members of the Autoplusiina. Convergent upon the condition found in Antoculeora,
the tegumen arms of Erythroplusia, in dorsal view, diverge at about 180 degrees, rather than forming the |
more usual inverted V-shape (104). Erythroplusia bears its clavus on a long, flattened peninsula (145), asin
Antoculeora and Macdunnoughia (s.1.). As in the latter genus (s.str.), the ventral sclerotization on the
aedeagus of Erythroplusia is restricted to a very thin, ventral band, convergent upon the condition seen in
the Argyrogrammatini. Finally, the form of the transtillae (155) is almost identical to that seen in
Sclerogenia.

In addition to the above states, the two constituent species of Erythroplusia show several other |
remarkable genitalic autapomorphies. The valves are asymmetrical (Fig. 378). The sacculus of the left
valve bears a distally directed, curved process, while that of the right valve bears a subapically recurved,

SYSTEMATICS OF THE PLUSIINAE 155

heavily-sclerotized ridge and no process. The corpora bursae of the females are highly simplified relative to
those seen in other autoplusiines. That of E. rutulifrons (Fig. 453) is a simple, cylindrical sac, somewhat
swollen where it recurves posteriorly through about 180 degrees. There is no obvious differentiation into
_ ductus and corpus bursae. The corpus bursae of E. pyropia is more swollen and tapers distally. In both, it is
extremely difficult to ascribe the terms apex and fundus with any degree of certainty (see also Characters:
209).
| a was originally proposed as a subgenus of Autographa (Ichinosé, 1962a; b). Subsequently,
it was elevated to full generic rank (Ichinosé, 1973) and included Antoculeora (q.v.) as a subgenus. The
precise relationships of Erythroplusia are still equivocal. There are some larval characters that suggest a
relationship with the Argyrogrammatini (a slender seta SD1 on segment A9 was used by Kljutshko, 1985a,
to associate the two taxa as sister-groups), while the chaetotaxy is identical in Erythroplusia and
Autographa (which included Sclerogenia; Ichinosé, 1973).

The larvae of both species have been described by Ichinosé (1962b). No foodplants were given for E.
rutulifrons but that of E. pyropia was stated to be Artemisia (Asteraceae). Miyata (1983) gave Arctium
(Asteraceae) and Oenanthe (Apiaceae) for both species. Erythroplusia is primarily north-eastern
Palaearctic in distribution, with the two species being broadly sympatric over much of their ranges.
However, E. pyropia extends further west than E. rutulifrons (Fig. 73), occurring throughout the southern
Himalayan foothills to Kashmir (Kostrowicki, 1961).

|

Clade 31 [113(1), 142(1), 176(1), 177(1)]. There are no unique and unreversed synapomorphies uniting the
seven constituent genera of clade 31. All four of the character states listed above are reversed in one or
more genera, and three also occur in the outgroups. Characters 176 and 177 have been discussed in detail
earlier; they are reversed in Loboplusia and Puriplusia respectively, and the former is also present in
Brachionycha. A saccular sheet (142) is absent in Loboplusia, while convergent developments are to be
found in Calophasia, Magusa and Lophoptera. Finally, the form of the vinculum (113) is interpreted as a
synapomorphy for clade 31. However, the character is subsequently reversed in the included clade 35, and
also occurs in Anagrapha. This, together with the uncertainty concerning the homology of states 1 and 2 of
142 and the ambiguity provided by Loboplusia (and to a degree Anagrapha), suggests that clade 31 may not
retain its current topology when further taxa and characters are analysed.

Clade 32 [99(2), 137(1), 139(1), 191(1)]. Allagrapha and Macdunnoughia (s.l.) are interpreted as
sister-taxa primarily on the basis of a produced tegumen apex (99). Also, the setae along the ventral edge of
the valve are borne on marginal prominences (137), a condition otherwise only seen in Omorphina, where
it is associated with a different valve shape. As noted earlier, clade 32 convergently shares a concealed
clavus base (139) with Rachiplusia. A relationship between Allagrapha and Macdunnoughia (s.1.) was first
suggested by Franclemont (1964).

Allagrapha [3(0), 16(1), 18(1), 58(0), 127(0), 201(1)]. Allagrapha is here interpreted as a monotypic genus
containing only A. aerea (Fig. 75) (McDunnough, 1944; Franclemont & Todd, 1983). Occasionally (e.g.
Eichlin & Cunningham, 1978), a second species, aereoides (misspelt in McDunnough’s (1944) figure
captions, Eichlin & Cunningham (1978) and Franclemont & Todd (1983) as aeroides), has been included.
Eichlin & Cunningham (1978) presented, as evidence for this inclusion, the absence in both species of a
raduloid on the larval hypopharynx (present in Diachrysia balluca), modifications of the male genitalia and
a strong superficial similarity. Ichinosé (1973), however, synonymized Allagrapha (as Agrapha) with
Diachrysia (as Plusia), noting a unique larval chaetotaxy (his type F) and the well-defined pocket in the
female segment A8, discounting the modification to the sacculus in A. aerea (Fig. 368).

However, while aereoides can remain in Diachrysia, I conclude that aerea is best retained for the present
in the separate genus Allagrapha, on the basis of the characters discussed in previous sections. Neverthe-
less, several Diachrysia species do complicate matters. D. leonina has a broad, cone-shaped vinculum
(113); D. nadeja (Oberthiir) bears a small, broad, triangular spine on the sacculus (141); and the marginal
setae on the valve of D. chrysitis, at least, are borne on small prominences (137). A more extensive analysis
of Diachrysia may eventually show that not only is this genus more closely related to some Autoplusiina
than to the Plusiina (as is concluded in the present analysis), but also that A/lagrapha is merely a divergent
species. However, it is unlikely that A. aerea and Diachrysia aereoides will prove to be sister-species.

In addition to the characters already mentioned, Allagrapha also has several genitalic autapomorphies.
The harpe is broad, flat and spoon-shaped; the clavus has a markedly swollen apex (Fig. 368); the LHS
basal sclerotization on the vesica bears very strong, sharp spines; and the ostium bursae is protected by a
large, spiculate lobe.

Allagrapha aerea is an undistinguished brown plusiine, showing, as already noted, strong superficial

156 IAN J. KITCHING

similarity in wing shape and pattern to Diachrysia aereoides, D. leonina and D. bieti. It is widely distributed
in southern Canada and the U.S.A., east of the Rocky Mountains (Eichlin & Cunningham, 1978). The
larvae are generally polyphagous on low plants (Forbes, 1954) such as Urtica (Crumb, 1956), Aster and
Glycine max (Papilionaceae; Eichlin & Cunningham, 1978).

Clade 33: Macdunnoughia (s.1.) [145(1), 150(1), 192(1), 206(1), 213(3)]. The two constituent taxa of
Macdunnoughia (s.1.), Macdunnoughia (s.str.) and Puriplusia, share two unique and unreversed synapo-
morphies: a harpe in the form of an inwardly curving hook (150) and the presence of spinose pads lateral to
the ostium bursae (192). Both characters are subsequently transformed, 150 in Macdunnoughia (s.str.) and
192 in Puriplusia. In addition, and as previously noted, clade 33 convergently shares the apomorphic states
of 145, 206 and 213 with Erythroplusia/Antoculeora, Rachiplusia and clade 35 respectively.

Puriplusia[177(0), 192(2),209(1)]. Puriplusia differs from Macdunnoughia (s.str.) primarily in the spinose
pads lateral to the ostium bursae being elongated into recurved grooves (192). In addition, the apex of the
corpus bursae has rotated through approximately 180 degrees and is thus directed anteriorly (209; see
Character analysis). Puriplusia also lacks concave lateral edges to the female St7 (177), probably as a
consequence of the reduced pleural membrane thickening (176). Autapomorphies not included in the data
set are the saccular sheet that is modified into an incurved hook and a sacculus bearing a distinct, blunt
spine (Fig. 369). The forewing pattern is also distinctive, although the high contrast two shades of grey seen
in typical P. purissima is replaced by an almost uniform grey-brown, on which the cross-lines are barely
distinguishable, in some forms of P. tetragona (Walker) (the synonymous P. semivitta (Moore); Dufay,
1977).

There are currently two species placed in Puriplusia, P. purissima (Fig. 77) and P. tetragona. P. semivitta
may prove to be a third or merely a colour form of P. tetragona. However, the complex is in need of revision
and there is at least one undescribed species in the BMNH collection. P. tetragona differs most markedly
from P. purissima in the presence of a row of short, curved cornuti on the vesica, similar to those found in
Macdunnoughia (s.str.) (see also Ronkay, 1982). Additionally, the recurved spiculate grooves lateral to
the ostium bursae are less well developed and there is no black dash above the Y-mark, distal to the
forewing discal cell.

P. purissima occurs in China, Korea and Japan (Hampson, 1913). The larva was described by Ichinosé
(1960) but no record of the larval foodplant was made. Miyata (1983) gave Artemisia and Arctium
(Asteraceae). P. tetragona is found further west and south in India. There are also specimens in the BMNH
collection from Taiwan. The undescribed species referred to above is exceptional in being collected from
west Sulawesi.

Macdunnoughia (s.str.) [150(2), 158(2), 175(1)]. Macdunnoughia (s.str.) is characterized in this data set by
the presence of an upward curving spine at the base of the harpe (150). It also bears a characteristic
sharp-edged auricular cornutus at the base of the vesica (175) and, like Erythroplusia and the Argyrogram-
matini, has the ventral sclerotization of the aedeagus restricted to a discrete narrow band (158).

Four species were placed in Macdunnoughia (s.str.) (type species: M. confusa) by Kostrowicki (1961).
Examination of the male genitalia of the holotype of M. camptosema Hampson showed that it lacks all the
apomorphies listed above for the Autoplusiina and correctly belongs in Autographa (comb. n.). The male
genitalia figured by Kostrowicki as A. camptosema (1961: fig. 54) are those of M. confusa (cf. his fig. 53). In
addition, Ichinosé (1973) has already pointed out that the female genitalia of M. crassisigna figured by
Kostrowicki are actually also those of M. confusa. Thus, Kostrowicki, to whom the male of M. crassisigna
and the female of A. camptosema were unknown, appears to have examined only one species of
Macdunnoughia (s.str.) (the appropriately named M. confusa!) but treated it as three separate species.
The last species transferred to Macdunnoughia by Kostrowicki was M. schalisema (Hampson), although
this time he did not examine any specimens. However, I have examined the genitalia of the female
holotype of M. schalisema and conclude, from its lack of both lateral spiculate pads near the ostium bursae
and a strongly ridged apex to the corpus bursae, that the species does not belong in Macdunnoughia.
Instead, a comparison with the types of Autographa emmetra Dufay shows schalisema to be extremely
similar both in forewing pattern and in female genitalia. I thus transfer M. schalisema to Autographa
(comb. n.) and synonymize Autographa emmetra (syn. n.) with it.

There are therefore only three species remaining in Macdunnoughia (s.str.): M. confusa (Fig. 76), M.
crassisigna and M. monosigna Chou & Lu. The last of these closely resembles the second in male genitalia
and may only be a colour form of M. crassisigna in which the basal half of the Y-mark has been lost. As
noted above (150: Character analysis), M. crassisigna has only a rudimentary upcurving spine at the base of
the harpe and is thus intermediate between M. confusa and Puriplusia. A more crucial feature, however, is

SYSTEMATICS OF THE PLUSIINAE 157

_ that in M. crassisigna, the apex of the corpus bursae is reflexed anteriorly and elongate, as in Puriplusia

__purissima. If this condition is interpreted as apomorphic, and a comparison with the other members of the
_ Autoplusiina suggests that it is, then a sister-group relationship between M. crassisigna and Puriplusia is

indicated, rendering Macdunnoughia (s.str.) paraphyletic.

M. confusa is a trans-Palaearctic species ranging from France, across central Europe and Asia, to
northern China, Japan and even the southern tip of Kamtchatka (Kostrowicki, 1961). However, until
about 1918, the species was only known from eastern Europe. After this date, M. confusa began a rapid
westward expansion of its range and is now probably resident in Italy, Spain and France. The species was

_ first recorded from Great Britain in 1951 (Dewick, 1951) but never became resident. It appeared to have

reached its limits in western Europe by 1955 (Bretherton et al. , 1983).
M. crassisigna occurs in China, the Korean peninsula and Japan, extending westward through northern

' Burma to India and Pakistan (Kostrowicki, 1961), while M. monosigna has only been recorded from

somewhere in China.

The larvae of both M. confusa and M. crassisigna have been described by Ichinosé (19586 and 19626
respectively). Both species appear to be somewhat polyphagous, but with a slight bias towards members of
the Asteraceae (Beck, 1960; Kostrowicki, 1961; Ichinosé, 19626; Miyata, 1983). Oord (1985) stated that
M. confusa larvae fed on Artemisia before hibernation and Lactuca and Matricaria after reawakening in

_ Spring.

Clade 34 [12(1), 59(0), 148(0)]. The three character states that unite the four constituent genera of clade 34
are very weak and it is unlikely that future studies will corroborate its monophyly. The main problem is
Loboplusia (q.v.) with its highly simplified male valves and numerous missing data. This genus is probably
not closely related to clade 35 (Sclerogenia, Autoplusia and Antoculeora).

Loboplusia[6(1), 141(1), 142(0), 144(0), 175(1), 176(0), 209(1)]. Loboplusia is perhaps most notable in the
present data set for the autoplusiine structures it does not possess. The genus lacks a broad saccular spine
(141), a saccular sheet (142), has a reduced clavus (144) (Fig. 371) and, while the lateral edges of St7 in the
female are probably concave (177), the pleural membrane is unmodified (176). The genus possesses an
auricular cornutus (175), but as the vesica of the slide preparation was not everted, its position on the vesica
is only assumed to be basal. Also, like Purip/usia, the apex of the corpus bursae is rotated to face anteriorly.

Nevertheless, the male Loboplusia has the most remarkable array of autapomorphic scent organs, their
diversity exceeding even Chrysodeixis. In addition to the usual sternal hair pencils, the hind tibia is
considerably swollen and bears a dense dorsal fringe of long, golden hairs. The hind tarsi are similarly
adorned but are of normal dimensions. The anal angle of the hindwing is extended into a long lobe
(superficially similar to the anal scent fold of swallowtail butterflies), which is folded and contains
numerous long hairs. Finally, the outer edge of the hindwing upperside and the basal costal edge of the
hindwing underside also bear modified scales. All in all, the male Loboplusia is a very distinctive and
unmistakable plusiine. In contrast, the female lacks all these features and has a similar appearance to an
Antoculeora, although without its golden scaling.

I know of only five specimens of the sole species contained in Loboplusia, L. vanderweelei (Fig. 74);
three males and two females. The holotype and paratype males were collected at 1700 m in west Java
(Roepke, 1941). Subsequently, Dufay (1973) described two females as a separate species, Autographa
pokhara. These were taken at 1500 m and 2000 m in Nepal and Burma respectively. The sexes were finally
associated by Dierl (1975), who reported a third male, which he had taken in Nepal at 2360 m, near
the locality of the holotype of A. pokhara. Little is known of the biology of the species, although Dierl
(1975) stated that it is only to be found in very wet mountain forests receiving over 6000 mm of rain per
year.

Clade 35 [113(0), 742(2), 198(1), 212(1), 213(3)]. The three genera of clade 35, Sclerogenia, Antoculeora
and Autoplusia are united by three good synapomorphies: the fusion of the saccular sheet with an overfold
of the costa (142); increased sclerotization posterior to the ostium bursae (198); and differential
sclerotization of the corpus bursae granulations (212). They also display strong sclerotized ridges on the
apex of the corpus bursae (213), a feature also seen in Macdunnoughia (s.1.).

Sclerogenia [95(0), 108(1), 155(1)]. Sclerogenia is another monotypic genus. The sole species, S. jessica
(Fig. 78), has several unique features of the genitalia not included in the current data set. The juxta has a
very pronounced basal bulge, while the fused saccular sheet/costal overfold forms an elongate, ventrally
curved process that just exceeds the bulged ventral edge of the valve (Fig. 361). The corpus bursae of the
female is distinctive (Fig. 455) in that the apex is reflexed and fused along the posterior section of the fundus

158 IAN J. KITCHING

and is entirely open to it internally, thus giving the appearance of two strigate ventral sacs. The remainder
of the fundus is characteristically lemon-shaped.

Sclerogenia was initially erected as a subgenus of Autographa by Ichinosé (1973) to receive the species
jessica, previously included in Macdunnoughia (Ichinosé, 1962b). Ichinosé (1973) still considered S. jessica
to be closely related to Macdunnoughia (then also a subgenus of Autographa), in view of the similarities of
the female genitalia. Sclerogenia was subsequently elevated to full generic status by Sugi (1982).

S. jessica is another species found in the eastern Palaearctic and southern Himalayan foothills. Hampson
(1913) recorded the species from Japan, Assam and the Punjab, while it also occurs in China (Chou & Lu,
1978). There is also a single specimen in the BMNH collection from northern Luzon in the Philippines. The
larva of S. jessica was described by Ichinosé (1960a). Its larval foodplants appear to be restricted to
members of the Asteraceae: Sonchus, Lactuca (Ichinosé, 1973), Ixeris and Arctium (Miyata, 1983).

Clade 36 [158(0), 198(2), 199(1), 200(1)]|. Antoculeora is interpreted as the sister-group of Autoplusia on
the basis of three synapomorphies: further sclerotization of the ventral area posterior to the ostium bursae
(198); a ventral origin of the anterior apophyses (199); and a distinctive form of antrum (200). In addition,
the aedeagus sclerotization of both genera is not restricted to a ventral band, instead the organ is generally
sclerotized, as in the presumed plesiomorphic state.

Antoculeora[18(1), 104(1), 116(0), 117(0), 778(2), 145(1), 189(0), 191(1)]. The outstanding features of the
genus Antoculeora are the highly modified and very distinctive genitalia. In the male, the whole of the
genital capsule appears to have moved posteriorly and to have rotated somewhat dorsally. This has
resulted in the pleurites losing contact with the dorsal edge of the tegumen arms, only retaining apical
contacts (118), and straightening (116) and narrowing (117) in the process. The tegumen arms are
extremely swollen anteriorly and, in dorsal view, diverge at 180 degrees (104), leaving a broad, U-shaped

anterior emargination. The valves have become irregularly and strongly digitate, taking the form ofastag’s |
antlers (Figs 376, 377). The tines are irregularly curved and asymmetrical, the left valve having three, the Me

right four, so that they partially interlock when the valves are closed. The saccular sheet has become a long,

apically serrate lance. The modifications in the female are less extreme but are doubtless partly correlated |
with those of the male. St7 is medially and posteriorly deeply emarginate, while the antrum has becomea |
heavily sclerotized, open, irregularly and asymmetrically lobed chamber (Figs 436, 437, 456). Together, |

they form a unique genital complex within the Plusiinae.

There are three, perhaps four, very closely related species in Antoculeora: A. ornatissima (Fig. 79), A.
locuples, A. lushanensis Chou & Lu and A. minor Chou & Lu, differing only in minor details of wing
pattern and valve shape. They may all eventually be referable to a single species. Initialiy, A. ornatissima

was placed by Kostrowicki (1961) in Argyrogramma. This, however, was based upon a misidentification |

(Ichinosé, 1963a; Dufay, 1970; Ichinosé, 1973). Ichinosé (19626) included ornatissima tentatively in
Macdunnoughia, then erected Antoculeora to accommodate it (Ichinosé, 1973). Antoculeora was first

treated as a subgenus of Erythroplusia but a relationship with Macdunnoughia was also suggested | —
(Ichinosé, 1973). This was formalized by Dufay (1977), who transferred ornatissima and locuples (as a |
synonym of the former) to Macdunnoughia. Antoculeora was resurrected as a full genus by Chou & Lu |—

(1979a) and so treated by Sugi (1982).

A. ornatissima has a similar distribution to Sclerogenia: Japan, Korea, China, the far eastern U.S.S.R. |

(Primorskiy Kray), Sikkim, Punjab, Nepal and Bengal (Dufay, 1977). A. minor and A. lushanensis are so
far only reported from China, while A. locuples is recorded in the BMNH collection from Japan and the
U.S.S.R (Primorskiy Kray). Nothing is known about the biology of this genus. Ichinosé (1973) stated that
the larvae had been reared on Sonchus and Lactuca but that the natural foodplant was unknown, while
Miyata (1983) gave Petasites. If these prove correct, then Antoculeora displays the same oligophagy,
restricted to the Asteraceae, as its close relative Sclerogenia.

Autoplusia [12(0), 14(2), 59(1), 64(1), 83(3), 131(0), 140(0), 216(1)]. Autoplusia is an exclusively New
World genus that is in need of revision. Four species are currently recognized: A. egena, A. olivacea, A.
egenoides Franclemont & Todd and A. illustrata (Guenée). To those can be added ‘Syngrapha’ gammoides
and I will limit my discussion to these five species. However, I understand from Mr F. Navarro that there
are several more species, currently languishing in either Syngrapha or ‘Plusia’ (s.1.) that are assignable to
Autoplusia.

A. egenoides was originally described by Hampson (1913) as the ‘much darker’ ‘ab. 1’ of A. egena. It was
named ‘ab. egenoides’ by Strand (1917), a name employed subsequently by Draudt (1940). For some
mysterious reason, it was elevated to full species by Franclemont & Todd (1983) in the latest North
American checklist. Even the very inclusion of A. egenoides in that list is odd for, as far as Iam aware, only

a

SYSTEMATICS OF THE PLUSIINAE 159

the holotype exists and that was collected in Mexico. Nevertheless, it would appear that egenoides has been
wrongly associated from the outset. I have examined the holotype and conclude that A. egenoides is
probably synonymous with A. olivacea. However, comparison of the genitalia is required to confirm this.

Autoplusia has a number of features in common with Rachiplusia, as noted above (characters 14, 64 and
216). It is also the only genus in the Plusiini to lack a saccular flange. In addition, as stated above, while the
type species, A. egena, lacks a broad, triangular saccular spine (Fig. 372) (141), this structure is present in
A. olivacea (Fig. 373) and ‘Syngrapha’ gammoides (Fig. 374). The autoplusias also possess a number of
} genitalic synapomorphies. The saccular sheet/costal overfold fusion is extended into a curved hook, the
. apex of the valve forming an oval lobe arising from the external surface of this structure. The sacculus is
extremely inflated. The everted vesica curves at about half its length through 90—200 degrees, the angle
varying between individuals within a species. On the outer surface of the angle, there is a large, distally
) directed, thorn-like cornutus, while the apical half of the vesica is covered with numerous small,
_ basally-directed cornuti, the arrangement of which is species specific. The distal part of the ductus bursae is
| heavily sclerotized, dorso-ventrally flattened and strigate, while the apex of the corpus bursae is elongated
| and apically recurved (Fig. 454). These features, with modifications, are present in A. egena, A. olivacea
) and ‘Syngrapha’ gammoides.
| A. illustrata, however, lacks all the valve apomorphies listed above but may have the thorn-like cornutus

(Eichlin & Cunningham, 1978: fig. 59). The female genitalia also appear to share few features in common
) with the other members of the genus (Eichlin & Cunningham, 1978: fig. 95), while close examination of the
| forewing pattern reveals only a superficial similarity. A. i//ustrata was only tentatively included in the genus
) by Kimball (1965) and although Eichlin & Cunningham (1978) stated that the male genitalia supported this
decision, I disagree. A detailed study of Autoplusia will probably show that A. illustrata (and the similar
‘Syngrapha’ egenella (Herrich-Schaffer)) ought to be removed from this genus. At present, I cannot
suggest a better placement.

‘Syngrapha’ gammoides (Figs 82, 83) isa Chilean endemic. Its life history and immature stages have been
described in detail by Angulo (1973) and Angulo & Weigert (1975). It is recorded as feeding on several
| crop plants in Chile and is an economically important pest in that country. A. egena is found from the
southern U.S.A. to Brazil and is also a pest species and high polyphagous (Eichlin & Cunningham, 1978).
It is commonly known in the U.S.A. as the Bean Leaf Skeletonizer and there have been occasional serious
outbreaks of this species (Kimball, 1965). The larvae have been described by Crumb (1956) and Eichlin &
Cunningham (1978).

The last species, A. olivacea (Fig. 81), was stated by Eichlin & Cunningham (1978) to be found only in
southern coastal California. However, in the course of this study, I found that there has been considerable
confusion between this species and A. egena (Fig. 80). The most obvious wing pattern difference concerns
| the form of the terminal line between the forewing veins R; and M3. In A. egena, it forms a distinct, single
pink crescent (Fig. 80), while in A. olivacea, the terminal line in this section of the wing is less distinct and is
| broken into three small crescents, one between each pair of veins (Fig. 81). Additionally, in A. egena, the
| forewing postmedial line is a distinct, thin, brown line, which is flanked on its outer edge by a broader,
pinkish brown band. It passes diagonally across the wing towards the apex but recurves basally just before
it meets the costal margin (Fig. 80). In A. olivacea, only the pinkish brown band is present. This becomes
very indistinct apically and does not recurve to meet the costa (Fig. 81). The Y-marks in both species are
very variable and of no use in identification. Those illustrated in Figs 80, 81 represent extreme forms that
occur in both species. These forewing pattern elements coincide with the much more distinctive genitalic
differences (see Eichlin & Cunningham, 1978). From this discovery, I draw three conclusions. First, the
specimens illustrated by Eichlin & Cunningham (1978: figs 168, 170) as A. olivacea and A. egena are both
A. olivacea. Second, A. olivacea is much more widespread than was thought. Specimens in the BMNH
collection have been taken in California, Mexico, Guatemala, Costa Rica and Colombia. Third, overall
forewing ground colour is of little use in distinguishing the two species. While A. olivacea, when fresh, is
distinctly darker and greener than A. egena, both fade to a similar pale brown. Incidentally, the holotype of
A. egenoides has the forewing pattern of A. olivacea, hence my suggestion above that the two species may
prove to be synonymous. The larva of A. olivacea has been described by Eichlin & Cunningham (1978)
and, like A. egena, is polyphagous.

Clade 37: Plusiina (83(1), 146(1)]. The constituent genera of the subtribe Plusiina are united solely by the
apomorphic state of character 146, a blade-like valve. As already noted above (Characters), this character
State is difficult to define but is readily recognized. The condition may have been misinterpreted in
Anagrapha (q.v.).

Diachrysia[3(0), 9(1), 16(1), 58(0), 65(Z), 165(1), 191(1)]. Ten species are currently placed in the Holarctic

160 IAN J. KITCHING

genus Diachrysia. Two, D. aereoides and D. balluca, are Nearctic, the remainder Palaearctic, with a
distinct concentration in Japan and neighbouring countries. As stated earlier, Diachrysia may not be
monophyletic, particularly with regard to Allagrapha, and thus the sole autapomorphy included in the data
set (65:1; the presence of a small number of large SD and D tarsal spines) may not be found much beyond
D. chryson and D. coreae Ogata).

There appear to be three species-groups within Diachrysia. The chryson species-group comprises five
species: D. aereoides, D. leonina, D. bieti, D. chryson (Fig. 84) and D. coreae. D. aereoides differs in being
a small, paler brown moth than the others. It is widespread in the northern U.S.A. and southern Canada,
extending south in the west to Colorado and in the east to Pennsylvania (Eichlin & Cunningham, 1978).
The larvae are recorded as feeding upon Mentha (Lamiaceae), Spiraea (Rosaceae) (Crumb, 1956) and
Aster (Forbes, 1954), and have been described by Crumb (1956) and Eichlin & Cunningham (1978). The
other four species are among the largest plusiines and have somewhat falcate forewings. D. bieti and D.
leonina are closely related and have brown forewings without any extensive metallic markings. The larva of
D. leonina has been illustrated by Kogi (1986). It bears enlarged conical dorsal pinacula, very similar to
those seen in the larva of Anadevidia. D. leonina is recorded as being polyphagous on such genera as
Urtica, Aster and Cirsium. D. coreae is similar to D. bieti and D. leonina but bears a diffuse, subapical,
quadrate, golden-green metallic patch on the forewing. It has long been regarded as synonymous with D.
chryson (e.g. Kostrowicki, 1961), in which the patch is much more discrete and intense. Nothing is
recorded of the biology of D. coreae but that of D. chryson is well known (e.g. Haggett, 1958; Bretherton et
al., 1983). The chaetotaxy has been described by Ichinosé (19626). D. chryson appears to be facultatively
monophagous on Eupatorium (Asteraceae) (e.g. Bretherton et al., 1983) but other foodplants have
been recorded: Salvia, Mentha (Lamiaceae) (Kostrowicki, 1961) and Arctium (Asteraceae) (Ichinosé,
1962b).

The zosimi species-group comprises only D. zosimi, and possibly D. balluca, although this latter species
also shows some resemblances to the chryson-group. D. zosimi has a unique colour pattern, in which the
forewings dorsal and distal of the discal cell are entirely metallic golden-green, with a ‘crushed velvet’
effect. The biology of this trans-Palaearctic species (Kostrowicki, 1961) has been described in detail by
Kasy (1953). The larvae are polyphagous but mostly on species of Rosaceae (Kasy, 1953; Kogi, 1986). D.
balluca has a similar forewing pattern to D. zosimi but has the large size and falcate forewing of the
chryson-group. It is restricted to the north-eastern U.S.A. and south-eastern Canada. The larval
chaetotaxy has been described by Eichlin & Cunningham (1978) and the foodplants have been given as
Humulus (Cannabinaceae) (Saunders, 1873) and Populus (Eichlin & Cunningham, 1978).

The third species-group, exemplified by D. chrysitis (Fig. 85), comprises three species in which the
forewing pattern consists of an H-shaped metallic patch, which may occasionally be represented by two
separate, transverse bands instead. D. stenochrysis (Warren) has a more diffuse pattern than the other two,
D. chrysitis and D. nadeja, and is restricted to the far eastern Palaearctic (Kostrowicki, 1961). The genitalia
of D. stenochrysis can be difficult to distinguish from those of D. chrysitis, especially the males (Ronkay,
1982). D. nadeja was also thought to occur only east of the Ob’ River basin, with isolated records as far west
as Bessarabia (Kostrowicki, 1961). However, like Macdunnoughia confusa, D. nadeja appears to be
undergoing a rapid westward expansion of range, having been recently discovered in Austria (Aistleitner
& Rezbanyai, 1982), Switzerland (Rezbanyai, 1980) and France (Dufay, 1986). If this progression
continues, then it may only be a matter of time before D. nadeja is captured in Great Britain and
Scandinavia. The life history has been described by Rezbanyai (1983), with that of D. chrysitis for
comparison. The natural foodplants of D. nadeja are unknown but Urtica and Plantago (Plantaginaceae)
were accepted in captivity (Rezbanyai, 1983).

D. chrysitis is a very variable species, as evidenced by no fewer than 14 varietal and aberrational names.
The species is widespread across the Palaearctic from Ireland to Japan. The immature stages have been
described in detail by Rezbanyai (1983) and numerous other accounts exist (e.g. Bretherton et al. , 1983).
The main larval foodplant is Urtica but many other species can be utilized.

There is also an occasionally recurrent taxonomic problem concerning D. chrysitis and the supposed
species D. tutti (Kostrowicki). D. tutti was described as a separate species from D. chrysitis (Kostrowicki,
1961) on the basis of smaller size and minor differences in forewing pattern and in the genitalia (Bretherton
et al., 1983). Lempke (1965) examined a long series of Dutch specimens and agreed that tnere were two
species, although there was also a great deal of intraspecific variation. However, Lempke (1966a)
subsequently re-examined some of his material and found that, with the aid of two pins, he could transform
the narrow ostium bursae [= antrum] of D. chrysitis into the broader ostium bursae of D. tutti. He thus
concluded that he had been wrong and that there was, after all, only one species, D. chrysitis, which was
subject to geographic variation (not unexpected given the extensive range of the species). Bretherton et al.
(1983) regarded tutti merely as an aberration of D. chrysitis. However, Rezbanyai (1985) has recently

— - ———<$<$<$£$_$$_—__—————_—_—_—
—————————————— ee —

SYSTEMATICS OF THE PLUSIINAE 161

suggested that fuiti may represent an eastern subspecies of D. chrysitis that is only partially reproductively
isolated from the nominate subspecies by incomplete pheromonal differences. The two taxa, which have
subsequently become partially sympatric, are thus currently in the process of hybridizing, further clouding
the distinction between them. The problem is obviously not yet fully resolved.

Clade 38 [105(1)]. This clade is supported by only a very homoplasious character state and is unlikely to be
corroborated by the results of future analyses.

Anagrapha [64(2), 113(1), 127(2), 143(2), 144(1), 189(1)]. Anagrapha is yet another of those genera that
shares putative synapomorphies with two widely separated groups of genera on the cladogram. The genus
is usually associated with Syngrapha (s.1.), due to the spined hind tibiae (64) and certain larval characters
(see below). They also share a juxta with a subapical process (127) and a T8 that is irregularly cut back
ventral to the base of the anterior apophysis (189). However, this latter character state also occurs in the
Autoplusiina, while a broad, flat, cone-like vinculum (113) further supports a relationship between some
members of this tribe and Anagrapha. Lastly, there is a remarkable similarity between Anagrapha and
Macdunnoughia (s.str.) (especially M. confusa) in forewing pattern (cf. Figs 86, 76 respectively).

However, Anagrapha shows none of the genitalic features characteristic of Macdunnoughia and its
relatives, although it has a number of autapomorphies. The harpe is bulbous basally and apically, and is
bent outwards at 90 degrees at about half its length. The valve is deeply excavated apically and the costa has
a fringe of dense, persistent, yellow hair. The costa terminates in a single distinct spine while the distal part
of the sacculus bears broad, curved setae, somewhat like those found in Ctenoplusia and its relatives. In the
female, the apex of the corpus bursae is recurved, as in many autoplusiines, but, while heavily scleroi.zed,
lacks the characteristic strong internal ridges of that subtribe (Fig. 459).

Anagrapha was erected by McDunnough (1944) to accommodate the species A. falcifera (Fig. 86). With
its markedly different genitalia, the species was considered to be a highly divergent branch of Syngrapha.
Eichlin & Cunningham (1978) agreed with this interpretation and synonymized Anagrapha with Syn-

_ grapha, tentatively deriving A. falcifera from the Syngrapha diasema lineage. One of the main reasons for

their placing of Anagrapha was the presence of seta SV2 on segment A1 of the larva. The feature was used
to unite Syngrapha, Anagrapha, Plusia and a branch of Autographa. However, while this seta also occurs in
Sclerogenia and Macdunnoughia (s.str.) (Ichinosé, 1962b), tantalizingly, it is absent in all other members
of the Autoplusiina known so far, and most notably in Puriplusia and Allagrapha. Thus, the relationships
of Anagrapha are as problematical using larval characters as with using adult characters and they are still
far from clear. Anagrapha was subsequently reinstated to full generic status by Franclemont & Todd
(1983).

A. falcifera is an abundant species throughout the U.S.A. and southern Canada (Eichlin & Cunning-
ham, 1978). No records are known for the countries south of the U.S.A. Commonly known as the Celery
Looper, the larvae sometimes cause economic losses to crops (Forbes, 1954). Overall, however, A.
falcifera appears to be of only minor agricultural importance.

Clade 39 [12(1), 108(1), 109(1)]. The most characteristic feature of clade 39 is the strongly S-shaped
vinculum, when seen in lateral view. However, this state is also weakly expressed in clade 21 of the
Euchalciina and further evidence is required to support the monophyly of clade 39.

Autographa gamma ([18(0)]. Autographa gamma (Fig. 87) was chosen to represent that branch of the genus
characterized by the lack of seta SV2 on segment A1 of the larva (Eichlin & Cunningham, 1978). It is thus
part of an overtly paraphyletic group of uncertain limits (cf. Trichoplusia). Eichlin & Cunningham
included ten species of Autographa in this group, divided into three lineages, depending upon the number
of processes on the ribs of the larval mandible.

The first lineage, A. precationis (Guenée) and A. bimaculata Stephens, was interpreted as the most
primitive because the processes on both ribs 2 and 3 had been retained. McDunnough (1944) noted that A.
Precationis was unusual and displayed a number of features in common with the Argyrogrammatini,
notably a bulbous coecum (157) and apical, basally-directed, setose scales on the valve (136).

The next most derived lineage had lost the process on the third mandibular rib and also comprised two
species, A. sansoni and A. rubida Ottolengui.

Those species that had lost both mandibular processes were considered to be the most derived and
included six Nearctic species: A. biloba, A. californica, A. pseudogamma (Grote), A. v-alba Ottolengui,
A. pasiphaeia (Grote) and A. flagellum (Walker). However, Neil (1984) has shown that the larva of A.
flagellum, which was unknown to Eichlin & Cunningham (1978), does possess a process on mandibular rib
2, and thus may be more closely related to A. sansoni and A. rubida.

162 IAN J. KITCHING

Of the Palaearctic species, five were examined from specimens in the BMNH collection, of which A.
gamma was found to lack SV2 on A1 (so too did A. bractea, see below).

Of the species for which life histories have been elucidated, most have been shown to be polyphagous on
herbaceous plants. A few (A. bimaculata, A. pasiphaeia, A. flagellum) have been recorded from only one
or two genera of larval foodplants (Eichlin & Cunningham, 1978), but this may reflect lack of information
rather than true oligophagy.

Two species are serious agricultural pests; A. gamma in the Palaearctic (Carter, 1984 and references
therein) and A. californica in the western Nearctic (Hyslop, 1912). A. biloba is an occasional pest of garden
crops (Forbes, 1954).

Clade 40 [6(0), 215(1)]. The members of clade 40 are united by the synapomorphic possession of a lateral
patch of thickened membrane on the corpus bursae (215). This feature is, however, lost in clade 44
(Syngrapha s.1.) and this must weaken the expectation that clade 40 will be corroborated by future studies.

Clade 41 [2/5(2)]. Autographa bractea and Cornutiplusia are united as sister-taxa by the further develop-
ment of the lateral thickened membrane on the corpus bursae into at least a shallow pouch (215). This
structure also occurs in the presumed close relatives of A. bractea (see below), A. aemula and A. excelsa
(Kretschmar) (Kostrowicki, 1961).

Autographa bractea [6(0), 12(0)]. A. bractea (Fig. 88) was originally chosen to represent the second
Autographa lineage comprising species that possess seta SV2 on larval segment Al. The choice was made
by picking the British equivalent of an American species (A. metallica (Grote)) included in the second
lineage by Eichlin & Cunningham (1978). However, they had not examined the larva of A. metallica and
when, subsequent to my decision, I examined the larva of A. bractea, I found that it actually lacked seta
SV2 on Al. A. bractea (and presumably those species, including A. metallica, that share the distinctive
forewing pattern; see below) is thus referable to the presumed plesiomorphic (paraphyletic) lineage of
Autographa exemplified by A. gamma. I will therefore discuss two further subgroups of Autographa, A.
bractea and its relatives, and those species known to possess seta SV2 on Al.

A. bractea could be considered to be part of a species-group in which the Y-mark has become modified
into an enlarged, single, silver or gold spangle. There are five species with such a pattern: A. bractea, A.
metallica, A. aemula, A. excelsa and A. kostjuki Kljutshko. The group is Holarctic in distribution, although
A. metallica is the only American representative. The life histories of three are known: A. aemula
(Hirschke, 1898), A. excelsa (Laasonen & Skvortsov, 1984) and A. bractea (e.g. Beck, 1960; Bretherton et
al., 1983). All three species are polyphagous but with a preference for the Asteraceae.

There are six species currently placed in Autographa that are known to have seta SV2 on segment A1: A.
corusca, A. mappa, A. ampla (Eichlin & Cunningham, 1978), A. nigrisigna (Walker) (Ichinosé, 1958a, b),
A. jota and A. pulchrina (Kitching, unpubl. data). The first three appear to feed as larvae primarily on
woody plants (Eichlin & Cunningham, 1978), although only A. ampla may be adequately recorded. While
the other three are polyphagous on herbaceous plants (Beck, 1960; South, 1961; Miyata, 1983), A. jota will
also feed on unspecified deciduous trees and shrubs (Bretherton et al., 1983). A. nigrisigna is an
agricultural pest in the eastern Palaearctic, where it is known as the Asiatic Common Looper or the
Cabbage Semilooper. There have been numerous studies of its bionomics (Ichinosé, 1959; 1960b; Ichinosé
& Asawa, 1964; Ichinosé & Shibuya, 1959), while the chaetotaxy of the larva must be the most
over-described of any plusiine, this having been presented in detail in at least three separate papers
(Mukerji & Singh, 1951; Ichinosé, 1960a; Betala & Kushwaha, 1972).

As was noted for Thysanoplusia, delimiting a group based on larval characters, when the larvae of only a
few species are known, results in many species possibly being wrongly excluded. There are several
autographas that fall into this category, and the elucidation of their life histories is an important
prerequisite to a comprehensive revision of the genus.

Cornutiplusia [64(1), 215(3)]. In Cornutiplusia, the lateral pouch of the corpus bursae mentioned above
(clade 41) is produced into a large pouch or into an appendix. The degree of development of this structure is
geographically variable and is the subject of another study (Kitching, in prep.). Cornutiplusia also has
spined hind tibiae and was thus associated with the genus Syngrapha (Hampson, 1913). Kostrowicki
(1961), however, thought the general appearance and genitalia suggested a closer link with Autographa.
Cornutiplusia stat. rev. is here treated as a good genus, certainly distinct from Syngrapha, perhaps less so
from Autographa (s.1.).

Cornutiplusia comprises a single species, C. circumflexa (Fig. 89), found throughout the more arid areas
of southern Europe, North Africa and western Asia, as far as Sri Lanka and Lake Balkhash (U.S.S.R.).

—

7
i
i
.
4
:
:

wo &

a

;
4
3

SYSTEMATICS OF THE PLUSIINAE 163

The species extends down the African rift valley to Kenya and Uganda, and there are isolated populations
on Grand Comoro and in South Africa/Namibia and the Gambia. The life history and larva have only been
very briefly described (Kirby, 1907). The larval foodplants have generally been stated to be members of the
Solanaceae (Kirby, 1907; Kostrowicki, 1961; Pinhey, 1975, who also gave Eschscholzia [Papaveraceae]).

Clade 42 [166(1)]. The three taxa that constitute clade 42 (Plusia, Lophoplusia and Syngrapha (s.1.)) are
united by their apparent shared possession of a grooved subapical cornutus. However, as discussed above
(648-step cladograms), the presence of such a structure has yet to be confirmed in Syngrapha (s.1.). The
species examined in this study all lacked a true subapical cornutus and were coded ‘not comparable’. The
results of future studies may therefore show that character 166 is actually a synapomorphy for Plusia and
Lophoplusia.

Plusia [12(0), 58(0), 83(3), 105(0)]. Plusia is a small genus containing at least five good species and a
number of other dubiously distinct taxa. All except one (P. venusta Walker) have a diagnostic forewing
pattern consisting of a yellow to deep orange ground colour, with the Y-mark modified into two large,
silver spangles and a white dash present in the cell M,, between the discal cell and the subterminal line (Fig.
90). In P. contexta Grote, these markings have become somewhat elongate, a transformation series that
culminates in P. venusta (Fig. 91). In this species, the Y-mark has become a thin, white streak, while the
ground colour of the forewing has become pale brown with cream margins. It thus superficially resembles
other noctuid genera such as Mythimna Ochsenheimer and Simyra Ochsenheimer. All species of Plusia
also have reduced dorsal abdominal tufting.

The species are all very similar in their genitalia. The males of P. venusta, P. putnami Grote and P.
nichollae (Hampson) cannot be reliably distinguished, while the females of the first two are similarly
virtually identical. P. festucae differs from P. putnami on minor points of genitalic structure, such as having
a larger clavus, a longer subapical cornutus that is pointed at both ends and a shorter ductus bursae
(Lempke, 1966; 1969; Dufay, 1969; Bretherton et al. , 1983). The males of P. contexta are unique in that
the valve costa terminates in a narrow, blunt projection (Eichlin & Cunningham, 1978). P. festucae and P.
nichollae differ from the other three species in having a somewhat round sclerotization on the corpus
bursae rather than one that is elongate.

The genus Plusia is Holarctic. P. festucae (Fig. 90) occurs across the Palaearctic region from the British
Isles to Japan. It is highly variable in size and wing pattern and this has led to the naming of numerous
aberrations, varieties and parochial subspecies of doubtful reality. P. putnami is even more widespread and
has a similar number of subspecific and infrasubspecific names. Four subspecies are currently recognized:
P. p. putnami occurs west across the northern U.S.A. and southern Canada to the Rocky Mountains but is
absent from the central states (cf. Pseudeva purpurigera) (Eichlin & Cunningham, 1978); P. p. gracilis
(Lempke) is restricted to western Europe (Dufay, 1969; Bretherton etal. , 1983); P. p. barbara (Warren) is
known only from a unique female holotype from Morocco; and P. p. festata Graeser occurs in the eastern
Palaearctic, from the Ob’ River basin to Japan. The remaining three species are all North American. P.
venusta and P. contexta have similar ranges to P. p. putnami, although neither extends as far west, while P.
nichollae is restricted to the west coast, from northern California to southern Alaska.

Larval foodplants are known for all species except P. nichollae. The genus Plusia feeds primarily on
grasses, sedges and other marsh and waterside monocotyledons (Beck, 1960; Dufay, 1969; Eichlin &
Cunningham, 1978; Bretherton et al., 1983; Miyata, 1983). The last group includes Sparganium (Sparga-
niaceae), Typha (Typhaceae), /ris (Iridaceae), Alisma (Alismataceae) and Juncus (Juncaceae). P. festucae
has also been reared on Crocosmia (Iridaceae), Salix, Brassica and Linum but these are species accepted in
captivity, rather than the natural foodplants. The larva shows several features unique in the Plusiinae that
are probably linked to its monocotyledonous diet. It is elongate, slender and cylindrical (Bretherton et al. ,
1983) and the head is about as wide as the rest of the body. The larva does not, therefore, taper anteriorly as
is typical of other plusiines. The crotchets are uniordinal, while the mandibles have a deep pocket on the
inner ventral edge that is apparently a modification for feeding on grasses (Eichlin & Cunningham, 1978).

- Pupation occurs in an elongate, strong, white cocoon, spun vertically on the underside of the leaves of its
foodplant and is placed near the apex so as to cause the leaf tip to droop and conceal it (South, 1961).

Clade 43 [59(0), 144(1), 166(2)]. There are no convincing synapomorphies at present to unite Lophoplusia
and Syngrapha (s.1.) as sister-groups. Characters 59 and 144 have been shown to be highly unreliable. The
possibility that a grooved or winged subapical cornutus (166) actually represents a synapomorphy for
Lophoplusia and Plusia, to the exclusion of Syngrapha (s.1.), has been discussed above.

Lophoplusia [18(0), 46(2), 47(1), 65(3), 107(1), 144(2)]. The genus Lophoplusia is endemic to the

164 IAN J. KITCHING

Hawaiian Islands. There are currently four described species: L. pterylota (Fig. 92), L. psectrocera, L.
violacea (Swezey) and L. giffardi (Swezey), with a possible fifth undescribed species represented in the
BMNH collection. The most readily recognizable feature of all species is the bipectinate antennae of both
sexes (46, 47) and this was the character state used by Zimmerman (1958) to define Lophoplusia when first
described. In addition, L. pterylota has numerous long, fine SD and D tarsal setae (65) and a single, dorsal
bar on the anal tube (107). L. giffardi shares the distinctively winged subapical cornutus (Zimmerman,
1958: fig. 303) but its presence in the other species requires confirmation.

Little is known about the biology of Lophoplusia. L. violacea and L. pterylota have recently been reared
on Stenogyne (Lamiaceae) and Clermontia (Lobeliaceae) respectively (K. Sattler, pers. comm.).

Clade 44: Syngrapha(s.1.) [64(1), 97(1), 127(2), 149(1), 162(0), 189(1), 191(1), 201(1), 215(0)]. Syngrapha
(s.1.) i.e. including Palaeographa and Caloplusia, but excluding Cornutiplusia and Anagrapha) is a large
genus of about 35 species. Its members are generally characterized by spined hind (and occasionally other)
tibiae but this character state also occurs in several other plusiine taxa, particularly if occasional spines are
counted (see Character analysis). The three taxa examined in this study also shared several other features
of interest. They are the only members of the Plusiini to have a straight uncus (97) and a sclerotized, sharply
pointed, hook-like harpe (149). Like Anagrapha, they have a subapical juxtal process (127) and a T8 cut
back ventrally below the base of the anterior apophysis (189). This feature also occurs in the Autoplusiina
(clade 30).

On Eichlin & Cunningham’s (1978) phylogenetic tree, Syngrapha occupied the apex, as the sister-group
of Plusia. They recognized a maximum of ten lineages within Syngrapha (excluding Anagrapha). I will
discuss four of these in this section, following the topology of Eichlin & Cunningham’s tree. S. surena
(Grote) will be discussed under Palaeographa; the celsa and rectangula species-groups under clade 45; and
Palaeographa, Syngrapha (s.str.) and Caloplusia in their own sections. More detailed descriptions of the
Nearctic species are given by McDunnough (1944) and Eichlin & Cunningham (1978).

S. ottolenguii was considered by Eichlin & Cunningham to be the most primitive Syngrapha. In many
ways, the species is halfway between this genus and Autographa. S. ottolenguii has the elongate, heavily
sclerotized antrum, the ventral edge of which projects beyond the ostium bursae (191; a feature seen in
Palaeographa and Caloplusia but not in any Autographa) and also spined hind tibiae (Eichlin &
Cunningham, 1978). However, the male genitalia are very similar to those of many Autographa species,
especially in having a long clavus (all other syngraphas have reduced clavi). Given that the form of the
female genitalia is likely to be apomorphic, while that of the male plesiomorphic, I will retain ottolenguii in
Syngrapha. McDunnough (1944) and Franclemont & Todd (1983) took a different view and placed the
species in Autographa. Either way, S. ottolenguii is clearly another ‘critical species’ and knowledge of its
immature stages is imperative. The species is restricted to Alaska and the Yukon Territory, west of the
Mackenzie River (Eichlin & Cunningham, 1978).

The next group proposed by Eichlin & Cunningham was a paraphyletic assemblage of all the remaining
syngraphas that lacked seta SV2 on larval segments TII and TIII. Three lineages (excluding Anagrapha)
were recognized within this group. The most primitive, in that its members retained the raduloid on the
larval hypopharynx, comprised four species: S. selecta (Walker), S. epigaea, S. altera (Ottolengui) and S.
octoscripta (Grote). The larvae of S. epigaea and S. selecta have been described by Eichlin & Cunningham
(1978). The former appears to be highly polyphagous on various trees and shrubs, while records for S.
selecta require confirmation due to confusion with the superficially very similar S. viridisigma (q.v.). Like
that species, S. selecta may also be shown to be restricted to conifers. S. octoscripta has been recorded as
feeding on Vaccinium. This group of syngraphas occurs across the northern U.S.A. and southern Canada.
S. altera is restricted to east of the Great Lakes; S. epigaea extends southward into Oregon, Wyoming and
Colorado; S. octoscripta ranges farther north to Alaska and Newfoundland; while S. selecta is so far known
reliably from scattered localities (Eichlin & Cunningham, 1978).

Of those species that lack both SV2 and TI and TIII and a raduloid, S. viridisigma was considered the
most primitive, due to genitalic and wing pattern similarities with the selecta species-group. As stated
above, this species is frequently confused with the rarer S. selecta but the genitalia of the two species are
distinct. The larva was described by Eichlin & Cunningham (1978) and various conifers were listed as larvai
foodplants. Some of these may, however, refer to S. selecta. S. viridisigma occurs throughout Canada, and
is found also in the far north-eastern U.S.A. and in the Rocky Mountains as far south as New Mexico.

The most advanced group of syngraphas lacking seta SV2 on TII and TIII (excluding Anagrapha) was
considered to be the diasema species-group. All these species have pale yellow hindwings, similarly
patterned forewings and show similarities in genitalic structure (Eichlin & Cunningham, 1978). The
included species were: S. diasema, S. snowii, S. sackenii, S. borea, S. diversigna and S. orophila. The larval
foodplants of only two species are known; Vaccinium (S. orophila; Crumb, 1956) and Betula and Populus

SYSTEMATICS OF THE PLUSIINAE 165

(S. diasema; Kostrowicki, 1961). S. diasema occurs throughout central Canada and southern Alaska
(Eichlin & Cunningham, 1978), and is also found in Scandinavia and the north-eastern U.S.S.R. It is
probably circumpolar in distribution (Kostrowicki, 1961). S. snowii is apparently endemic to New Mexico,
while S. borea is found in the far north of Canada and in Greenland. S. diversigna occurs in the Canadian
Rockies and southern Alaska, and is broadly sympatric with S. orophila, which extends further south to
Wyoming. S. sackenii is found on the eastern slopes of the Rocky Mountains in the U.S.A. and Alberta,
and appears to be sympatric with S. orophila in that part of the latter’s range in which S. diversigna is absent
(Eichlin & Cunningham, 1978).

Palaeographa [6(1)]. When Palaeographa was first described as a subgenus of Syngrapha (Kljutshko,
1983), four species were included: S. interrogationis, S. alpina Ichinosé, S. gilarovi Kljutshko and S.
transbaikalensis (Staudinger). S. alpina had previously been synonymized with S. nyiwonis Matsumura
(Sugi, 1982), and this latter taxon is also referable to Palaeographa. Other candidates include S. pyrenaica
Hampson, S. flammifera Chou & Lu and S. u-aureum (Guenée). These last three species have all, at one
time or another, been considered synonyms of S. interrogationis. S. transbaikalensis apparently grades into
S. interrogationis in Korea (Ronkay, 1982) and may thus be yet another form of the latter.

As for the previous major Syngrapha subgroup, those species that possess seta SV2 on segments TII and
TIII were also subdivided by Eichlin & Cunningham (1978) based upon the presence or absence of the
raduloid. S. interrogationis and S. u-aureum (together with S. surena, see below) constituted that lineage in
which the raduloid was retained and was therefore considered primitive. This taxon may correspond to
Palaeographa but confirmation of the presence of the raduloid in the larvae of the other species is required.
Also, at present, well-documented synapomorphies for the members of Palaeographa are needed to
corroborate the monophyly of the subgenus.

S. interrogationis (Fig. 93) is a widespread Holarctic species. It occurs in northern Canada and Alaska
(Eichlin & Cunningham, 1978), throughout northern and western Europe (the British Isles, Iceland,
Scandinavia) and east through central U.S.S.R. to the Pacific seaboard. S. pyrenaica is endemic to the
Pyrenees and may be no more than a local race of S. interrogationis. S. transbaikalensis, as its name
suggests, occurs in the north-eastern Palaearctic region, east of Lake Baikal to Japan and Kamtchatka. S.
nyiwonis is found in Japan and the neighbouring land areas of Sakhalin and the Kurile Islands (Ichinosé,
1963b). Further details of the distribution and ecology of ‘S. alpina’ were provided by Kljutshko (1983),
who also stated that S. gilarovi was restricted to the area of Kamtchatka. S. flammifera was described as an
aberration of S. interrogationis by Hoyningen-Heune (1901) but was subsequently elevated to full species
level, from subspecies (sic), by Chou & Lu (1979b). Given that the collection locality of the original
specimen of flammifera was Estonia and Chou & Lu presumably examined specimen(s) from China, then
either S. flammifera is a very widespread (or disjunct) but rare and highly undercollected species or, as is
generally considered, is merely a form of S. interrogationis. The diagnostic characters used by Chou & Lu
(1979b) (‘transverse cordate stigma of [the] fore-wing and [the] truncate apex of the clavus’) suggest that
the latter hypothesis is most likely correct. S. u-aureum occurs throughout Canada and the north-eastern
U.S.A., with an isolated record from Colorado (Eichlin & Cunningham, 1978).

The life history of only one species is known, European S. interrogationis (Bretherton etal. , 1983). These
authors give Calluna and Vaccinium (Ericaceae) as larval foodplants, while Kostrowicki (1961) added
Erica and Urtica. S. u-aureum is said to feed on Vaccinium (Forbes, 1954) but no further details of its
biology are known.

A second species, S. surena, was associated with S. interrogationis by Eichlin & Cunningham (1978) on
the basis of similar genitalia. S. surena is very rare, Hampson (1913) never saw it and Eichlin &
Cunningham (1978) knew of less than a dozen specimens. It occurs in Newfoundland, Ontario, Quebec
and Maine.

Clade 45 [7(1), 24(1), 127(3)]. Three synapomorphic character states unite Syngrapha (s.str.) and
Caloplusia as sister-groups. All members of both taxa (see below) have yellow hindwings (7) except S.

_parilis, in which they are white. A juxta with a subapical, sharp spine (127) is also almost universal but is

absent in S. montana. Small eyes with a prominent rim (24) are restricted to a subset of potential members
of clade 45; Caloplusia, Syngrapha devergens and S. ignea. All three of these conditions are absent in the
putatively related ce/sa and rectangula species-groups, which I will now discuss, separately from Syngrapha
(s.str.) and Caloplusia.

Eichlin & Cunningham (1978) considered as most highly evolved, that branch of Syngrapha in which seta
$V2 was present on segments TII and TIII of the larva but the raduloid had been lost. Four lineages were
recognized within this group. Caloplusia (q.v.), with spines on all the tibiae, was interpreted as the first
lineage to diverge. Next was the rectangula species-group, which was considered to be closely related based

166 IAN J. KITCHING

on similarities in larval and genitalic structure. The celsa species-group formed the sister-group of
Syngrapha (s.str.). Thus clade 45 was treated by Eichlin & Cunningham (1978) as polyphyletic (Farris,
1974), and the three synapomorphies listed above would have to be interpreted as convergent develop-
ments. Further study of Syngrapha is required to clarify this issue.

The celsa species-group comprises only S. celsa and S. angulidens. Both species feed as larvae on various
conifers (Crumb, 1956; Ross & Evans, 1957; Prentice, 1962). Nothing further is known of the biology of S.
angulidens but the larva of S. celsa was described by Crumb (1956) and Eichlin & Cunningham (1978). S.
celsa is restricted to the western slopes of the Rocky Mountains from Alaska to southern California, while
S. angulidens is found on the eastern slopes, from southern Canada to Arizona. There are also records of S.
angulidens from New Hampshire and Nova Scotia in the east (Eichlin & Cunningham, 1978).

The rectangula species-group comprises four species: S. rectangula, S. alias (Ottolengui), S. abstrusa
Eichlin & Cunningham and S. cryptica Eichlin & Cunningham. The latter two species have only recently
been separated from S. alias (Eichlin & Cunningham, 1978). Nothing is known about their biology but
some of the records reported for S. alias may well apply to these species. Like the celsa-group, S. rectangula
and S. alias feed as larvae on diverse species of conifers (Crumb, 1956; Ross & Evans, 1957; Prentice, 1962;
Tietz, 1972). This habit was, together with the presence of SV2 on TI and TIIL, sufficient for Crumb (1956)
to consider proposing a separate genus (genus no. 11). However, the setal character is now known to be
much more widespread and so only monophagy on conifers remains as a putative synapomorphy for the
rectangula and celsa species-groups. S. alias and S. rectangula are found across the northern U.S.A. and
southern Canada, although the former also extends along the Rocky Mountains north to the Yukon
Territory and south to Arizona. The distributions of the other two species are likely to be incomplete, due
to confusion with S. alias. S. cryptica has so far been recorded from Wisconsin, Michigan and New
Hampshire, and S. abstrusa additionally from Ontario and Maine (Eichlin & Cunningham, 1978).

Syngrapha (s.str.) [83(0) |/Caloplusia [59(1), 63(2), 64(2)]. I will deal with these two taxa together because
the distinction between them has been blurred in recent years. Caloplusia was originally erected by Smith
(1884) for those northern species of ‘Plusia’ (s.1.) with ovate eyes and spines on all the tibiae. However,
there were no explicitly included species until C. hochenwarthi and ‘C. devergens’ were placed in the genus
some seven years later (Smith, 1891). Warren (1913) designated C. hochenwarthi as type species in April
1913 (Nye, 1975) and added C. tibetana (Staudinger) and C. composita Warren. Hampson (1913;
November) subsequently employed the same type species and added ‘C. virgula’ (now Rachiplusia), C.
alticola (Walker) and C. ignea (Grote). In both works, Caloplusia was distinguished from Syngrapha by the
possession of spines on all the tibiae, rather than on just the hind tibiae (characters 63 and 64). Both authors
also incorrectly cited Syngrapha ain (Hochenwarth) as the type species of that genus and thus S. devergens
could be comfortably accommodated in Caloplusia.

This error was corrected by McDunnough (1944) and devergens (together with alticola) was transferred
to Syngrapha. McDunnough’s discovery that neither of these two species had the extensive tibial spining
claimed for them (an observation confirmed in the present study) allowed him to leave Caloplusia as a
distinct genus. The separation was retained by Forbes (1954), Crumb (1956) and Kostrowicki (1961). The
last author committed the same error as Hampson (1913) but on this occasion it was S. interrogationis that
was mistakenly treated as the type species of Syngrapha.

Eichlin & Cunningham (1978) held the opinion that the degree of tibial spining was not a character of
genus-level significance and synonymized Caloplusia with Syngrapha. However, only ignea was explicitly
transferred; hochenwarthi was included only by default. Franclemont & Todd (1983) followed this
synonymy and also sank the Nearctic S. alticola to the Palaearctic S. devergens.

There are thus eight species in clade 45 as constituted in this study. For convenience, I shall retain the
names Caloplusia and Syngrapha (s.str.). The latter includes S. devergens (Fig. 94), S. montana, S.
microgamma, S. parilis (Fig. 96) (following Eichlin & Cunningham, 1978) and S. ain. The last species has |
never previously had its relationships within Syngrapha analysed. An examination of the male genitalia of |
S. ain figured by Ichinosé (19625) showed that it has the same general facies and large, basal, strongly |
curved cornutus on the vesica as §. microgamma, and it would seem that the two may be sister-species.
Caloplusia comprises C. hochenwarthi (Fig. 95), C. composita and C. ignea.

The biology of none of the Nearctic species of Caloplusia or Syngrapha (s.str.) is known, nor of the |
Nearctic populations of the Holarctic species. In Europe, S. ain is recorded as feeding as a larva on Larix |
(Pinaceae); S. microgamma on Salix (Salicaceae); S. devergens on Silene (Caryophyllaceae), Viola |
(Violaceae), Plantago (Plantaginaceae) and Geum (Rosaceae); and C. hochenwarthi on various umbellif- |
ers (Kostrowicki, 1961). Tietz (1972) added numerous arctic-alpine plants to the list for C. hochenwarthi |
(Antennaria, Carex, Castilleja, Gentiana, Penstemon, Salix and Vaccinium). The larva of C. hochenwarthi |
has been briefly described by Kirby (1889), but the chaetotaxy of all the species remains virtually unknown.

;

|
;

SYSTEMATICS OF THE PLUSIINAE 167

A larva of S. ain examined in the course of this study proved to possess seta SV2 on segments TI and TIII.

S. montana is the only wholly Nearctic species of Syngrapha (s.str.). It occurs mostly in the north-eastern
comer of the U.S.A. and adjacent areas of Canada south of the St Lawrence river. There are isolated
records from Wisconsin and Saskatchewan. The other endemic Nearctic species is Caloplusia ignea, found
mainly in the Rocky Mountains from Alaska to California and Colorado, with additional records across
northern Canada (Eichlin & Cunningham, 1978). C. composita is known only from the mountains
south-east of Lake Balkhash (U.S.S.R.), while C. hochenwarthi has a disjunct distribution in Scandinavia,
the European Alps, Macedonia, and various localities throughout the U.S.S.R. from Armenia and the
Caucasus to the Amur river and Kamtchatka (Kostrowicki, 1961). Syngrapha ain occurs in the mountains
of central Europe and in the central U.S.S.R. from Lake Balkhash to the River Lena. There are also
records from the Amur river (Kostrowicki, 1961).

The remaining three species are Holarctic. S. microgamma is present throughout central Europe,
Scandinavia and along the Baltic coastline, with scattered records in the central U.S.S.R. and Kamtchatka,
and throughout Canada and the northern U.S.A., with an isolated record from Colorado. S. devergens also
occurs in the European Alps, Scandinavia and south-east of Lake Balkhash and in the Nearctic, extends
throughout the Rocky Mountains from Alaska to Colorado, and across to Hudson Bay. S. parilis is
circumpolar at high latitudes in Canada, Alaska, Greenland, Scandinavia, the Arctic Ocean coasts and
Novaja Zemlya (U.S.S.R.) (Eichlin & Cunningham, 1978; Kostrowicki, 1961). Its absence from localities
further east is almost certainly due to lack of collecting in these areas.

Intra-outgroup analysis

The cladistic relationships between the outgroup taxa are not of primary importance in this study.
However, several interesting patterns were found and it might be useful, for future studies, to discuss them
briefly. I will only deal with six of the possible ten outgroup clades.

The complement of clade 51 (38(0), 40(0), 48(1), 56(7), 178(1)]. If the cladogram shown in Fig. 27 is
rerooted on the branch subtending clade 51, then a new clade is formed (the complement of 51) that
hypothesises that the sister-group of the plusiines is clade 47 (Magusa + the stictopterines + the
euteliines). As might be expected, given that this topology is only one of several possible (see 648-step
cladograms), the data supporting this sister-group relationship are relatively weak. The Plusiinae and the
members of clade 47 are unique only in possessing a broad marginal flange to the dorsal tegular arm (56).
However, as noted earlier, the states of this character are not clear cut and may be subject to
reinterpretation. A better character is the possession of a pleural pouch (178), although this must be
interpreted as having been lost in Magusa.

Thus, although clade 47 is a reasonable candidate for the sister-group of the Plusiinae, the evidence is far
from unequivocal. Further analysis is necessary, including a detailed study of the relationships of the taxa
comprising clade 47, before the sister-group of the Plusiinae can be named with any degree of certainty.

Clade 47 [20(2) , 22(0), 23(1), 26(1), 34(1), 37(1), 81(0), 82(1), 83(3), 107(1), 118(0), 156(1), 210(1)]. There
are seven good synapomorphies to unite the Stictopterinae, Euteliinae and Magusa into a monophyletic
group: scale bases on the wings arranged into regular transverse lines (20); bulbous eyes and a prominent
frons (23, 26); the setae on the proboscis continuing almost to the apex (34); a smooth ventral edge to the
basal two segments of the labial palps (37); and the form of the hair pencils (81) and St8 (82). In addition,
the apex of the pleurite is recurved over the tegumen arm to articulate in a pocket on the inner edge (118;
also seen in Calophasia); the aedeagus is very small compared to the rest of the genitalia (156; reversed in
Paectes); and there are regular ridges on the corpus bursae (210; present also in Calophasia and Cucullia).

However, the monophyly of clade 47 should not be accepted uncritically, given the paucity of included
genera examined. Although Magusa and Stictoptera look very alike superficially, experience of similar
Situations in the Plusiinae (e.g. Anagrapha and Macdunnoughia confusa) suggests that this may be a poor

guide to close cladistic relatedness. The genitalia of Stictoptera and Magusa in both sexes are very different.

Also, Magusa is a trifine, with rudiments of a trifine hair pencil (75), whereas the stictopterines and
euteliines are unequivocally quadrifine with no trace of trifine hair pencil components. Although these
characters may be prone to frequent reversal and convergence, they do suggest that the support for clade 47
may not be as convincing as it may first appear. Also, the sternal hair pencils of Magusa are markedly more
complex than any seen in the Stictopterinae and Euteliinae, and studies to confirm the homologies
proposed in the current investigation are required. Thus, while there can be little doubt that the
Stictopterinae and Euteliinae are sister-taxa (see below), the conclusion that Magusa is their sister-group
must still be considered open.

168 IAN J. KITCHING

Clade 48 [13(0), 14(3), 15(0), 17(1), 18(0), 25(1), 43(1), 52(2), 57(1), 62(1), 66(1), 68(1), 70(1), 72(1),
73(3), 114(1), 123(1), 153(1), 161(1), 179(1), 180(1), 190(1), 195(1), 202(1), 211(1)]. Next to the
Abrostolini, clade 48, the Euteliinae + Stictopterinae, is the best corroborated clade found in this study.
There are no fewer than 11 unique and unreversed synapomorphies to support it, together with a further
nine in which the apomorphic state has been subsequently lost in one of the included genera and another
four where a convergent development has been recorded in only a single genus outside the clade.

The euteliines and stictopterines share the following good synapomorphies: a reduced female frenulum
(14), a full quadrifine hindwing venation (15; probably synapomorphic only in the context of the taxa
included in this study, but, given Magusa is trifine, maybe a reversal); a frons scaled only at the edge (25); a
patagium with a drawn-out attachment point (57); weak SV and V hind basitarsal spines (62); hindlegs
shorter than the midlegs (66); dorsal furcal arms that are curved posteriorly (70); the form of the internal
flange on T2 (73); ventrally or subventrally positioned valves (153); a small, oval plate in the ductus
ejaculatoris (161); and anal papillae in which the inner surfaces are turned to face at least partially
posteriorly (179). In addition, the Stictopterinae show partially-developed ‘wings’ on the proboscis
styloconic sensilla (33) (Fig. 113). If this condition is interpreted as intermediate between the plesiomor-
phic state 1 (Fig. 112) and the apomorphic state 2 (Figs 114-116), then it further corroborates the
monophyly of both clade 48 and the Euteliinae.

They also share the following putative synapomorphies but these are either lost in one of the constituent
taxa (indicated by ‘— genus’ in parentheses) or occur in another genus outside clade 48 (‘+ genus’ in
parentheses). The character states involved are: underside of the forewing cell without setose scales (13, +
Chalcopasta); forewing vein R,+R3+R, markedly converging on R,; then curving away distally (17; —
Paectes); vom Rath’s organ conspicuously papillate (43; — Anuga); ventral arm of the tegula spatulate (52;
— Stictoptera); hindleg inner apical spurs approximately three-quarters length of basitarsus (68; —
Lophoptera); abdomen distinctly tapered posteriorly (72; — Lophoptera); dorsal and ventral flanges of
vinculum arms fused across to give a flattened, cone-shaped structure (114; + Rachiplusia); vinculum and
tegumen broadly joined, pleurite completely fused to tegumen (123; + Diloba); inner surface of the anal
papillae densely setose (180; + Calophasia); anterior apophyses absent (190; — Stictoptera); female T8
ventrally produced anteriorly to lie alongside antrum dorsally and exceeding it (195; — Stictoptera); female
T8 represented by a thin band dorsally (202; — Stictoptera); and the corpus bursae with an invaginated
signum (211; — Anuga).

Examination of the preceding 13 characters reveals that there is no strong pattern to the homoplasies
that would imply that either subfamily is paraphyletic relative to the other. It could be argued that
Lophoptera is more closely related to the euteliines than it is to Stictoptera, because it shares four putative
synapomorphic features missing in Stictoptera (52, 190, 195, 202). However, there are two character states
(68, 72) that imply the opposite relationship and, given that there are other characters to corroborate the
monophyly of the two subfamilies, the six states just mentioned must be interpreted as homoplasies.
Certainly, no firm conclusions should be drawn about the internal structure of clade 48 when only 4 of the
40 or so generic-level taxa have been examined, but equally, I think it is unlikely that the monophyly of the
Stictopterinae and of the Euteliinae will be refuted.

Although Hampson (1912) recognized the reduced female frenulum in the Stictopterinae and Eute-
liinae, he misinterpreted its structure in the latter group, referring to it as being ‘single’. He considered the
two subfamilies to be parallel developments from the ‘Ophiderinae’. Seitz (1937) also observed that the
euteliines and stictopterines had a number of features in common, including the large, bulbous eyes (23),
from which he concluded that they were closely related.

The recognition of a close relationship suffered a setback with Richards’ (1932) study of the tympanal
organs of noctuids. He observed a secondary tympanal hood (pleural pouch) in the stictopterines and the
plusiines and derived the latter from the former. The euteliines were thought to have much in common with
the acontiines and although a membranous secondary tympanal hood was observed in Eutelia, the
connection with the Stictopterinae (and Plusiinae) was not made. Forbes (1954) referred only to the
stictopterine-plusiine relationship, suggesting that these two subfamilies might perhaps be combined, and
the euteliine-stictopterine connection was lost.

Recently, in a study of several Bornean noctuid subfamilies, Holloway (1985) resurrected the hypothesis
of a euteliine-stictopterine connection when he noted the presence of a small, oval sclerotization in the
ductus ejaculatoris (character 161) of the members of these two subfamilies. He also noted several other
interesting similarities (e.g. the morphology of the anal papillae) and these formed the basis of the
characters employed in this study.

Clade 49: the Stictopterinae [8(/), 14(5), 61(1), 62(2), 121(1), 151(1), 181(1)]. Five unique and unreversed
synapomorphies were found for the two stictopterine genera included in the data set. These are: a hyaline

en a

SYSTEMATICS OF THE PLUSIINAE 169

hind wing base (8); a female frenulum reduced to a single seta (14); suppressed SV and V spines on the
basal half of the hindbasitarsus (62); a deeply cleft valve (151); and apically ‘frayed’ anal papillae (181). Of
these, not all stictopterines show the apomorphic states of characters 8 (e.g. Odontodes) and 151 (e.g. the
Lophoptera khasiana Warren species-group) (Holloway, 1985). Character 62 requires confirmation from
the other genera but Holloway (1985) found all Bornean stictopterines to have the apomorphic states of
characters 14 and 181. To these he added three more putative synapomorphies: the form of the tympanal
hood, the general wing facies and a distinctive Y-shaped pupal cremaster (Gardner, 1948b). The
monophyly of the subfamily thus seems reasonably well corroborated.

Holloway also provided brief reviews of the morphology (both of adults and immatures), biology and
zoogeography of the subfamily. In particular, he concluded that the two major lineages of stictopterines
have different larval foodplant preferences. The Stictoptera genus-group appears to be restricted to
members of the Clusiaceae (Guttiferae), while the Lophoptera genus-group is primarily recorded from the
Dipterocarpaceae. According to Thorne (1976) and Heywood (1978), these two families are closely
related, both being members of the Theales, but more recent classifications of the angiosperms (Dahlgren,
1983; Thorne, 1983) disagree, placing the dipterocarps in the Malvales. Members of the Stictoptera
genus-group may be of potential economic importance as defoliators of crop trees such as Garcinia
(mangosteen) (Holloway, 1985).

Clade 50: the Euteliinae [33(2), 35(2), 36(1), 46(1), 47(0), 58(1), 68(2), 71(1), 74(1), 107(0), 115(1), 122(1),
129(1), 148(0), 152(1), 157(1)]. The Euteliinae are even better defined and more autapomorphic than their
sister-taxon, with no fewer than ten unique and unreversed synapomorphies to corroborate their
monophyly. Only Paectes and Anuga were found to possess the following features: winged styloconic
sensilla on the proboscis (33) (but see clade 48 above), which is nodular only at the extreme apex (35) and
has regular, dense, basal rows of microspines (36); asymmetrical, bipectinate male antennae (46), with the
pectinations bearing only a single, small, apical seta (47); hindleg inner apical spurs that equal or exceed
the hind basitarsus in length (68); two internal flanges on St2 (74); vinculum arms that are distally parallel
or convergent (115); socii (122); a crescent- or boomerang-shaped, weakly sclerotized juxta (129); and a
somewhat trapezoid valve (152).

Of these, however, the antennae are ciliate or filiform in Anigraea Walker and Targalla Walker
respectively; the valves of other genera are more varied in shape; and socii are frequently absent
(Holloway, 1985). Many of the other characters listed above require confirmation but all genera appear to
possess the flanges on St2, which may be associated with the characteristic resting position of the adults
(Holloway, 1985; Kitching, 1984).

Reviews of the morphology of the adults and immatures, biology and zoogeography of the Euteliinae
have been presented by Holloway (1985). The lack of a pupal cremaster and a bias towards the
Anacardiaceae as larval foodplants (but see also Peigler, 1986) may also represent potential apomorphies
for the subfamily. Several genera have been recorded as agricultural pests, especially of mango (Man-
gifera).

Holloway (1985) did not propose a cladistic hypothesis for the euteliine genera. Mell (1943) recognized
four genus-groups based on Penicillaria Guenée (as Bombotelia Hampson), Eutelia Hibner (as Phlo-
gophora Treitschke), Anuga/Mimanuga Warren and Paectes, but only examined a very small number of
species. Thus, the internal relationships of the Euteliinae remain to be elucidated.

Clade 55 [20(0), 27(1), 31(0), 32(2), 34(2), 46(2), 55(1), 71(1), 96(1), 208(1)]. The position of the enigmatic
genus Diloba, and its single species, D. caeruleocephala, has changed with almost monotonous regularity. I
gave a brief historical review (Kitching, 1984), charting its course from the Bombyces, via the Noctuidae:
Bombyciformes, Notodontidae: Dilobinae, Noctuidae: Plusiinae, Thyatiridae, Noctuidae: Acronictinae,
Noctuidae: Pantheinae, Lymantriidae and Dilobidae, to the Noctuidae: Cuculliinae. Diloba has certainly
been around!

Possible inclusion in the Cuculliinae was first mooted by Tams (quoted in Varley, 1962), who suggested a

telationship with the American genus Psaphida Walker. Minet (1982), in a critique of Kiriakoff’s (1970)

work on the tympanal organ of Diloba, independently hypothesised a place for the genus in the
Cuculliinae, associating it with Episema Ochsenheimer. In this study, I decided to follow up Tams’
Suggestion to test the relationships of Diloba. The members of the cuculliine tribe Psaphidini are mostly
Nearctic and poorly represented in the BMNH collection. However, specimens of the Holarctic genus
Brachionycha are more numerous, and this taxon was chosen to represent the Psaphidini, although there
have been reservations concerning its placement in this tribe (Forbes, 1954).

I found six putative synapomorphies that were unique to Brachionycha and Diloba. These would appear
to confirm the place of the latter in both the Noctuidae and the Cuculliinae. However, a more extensive

170 IAN J. KITCHING

investigation is required to confirm the relationship of these two genera with the Psaphidini and/or
Episema. The synapomorphies are: the presence of a vestigial remnant of forewing vein CuP (21); a
reduced, non-functional proboscis with no distal styloconic sensilla and setae continuing to the apex (32,
31, 34); the ventral arm of the tegula produced on a lateral extension of the main body (55); and an
apparently small corpus bursae, but with extremely elastic walls (208).

Some of thse conditions are somewhat unconvincing as synapomorphies (especially the reduction
characters) but the unusual tegulae have not so far been found elsewhere. There are numerous differences
between the two genera but it was not expected that Brachionycha would prove to be the sister-group of
Diloba. However, I hope that I have shown that future studies of the relationships of Diloba should
concentrate on the Psaphidini and not on the supposed notodontid connection.

Consensus classification

From the preceding evaluation of the preferred 648-step cladogram, it is clear that there are both clades
and characters of highly dubious value. For example, the apomorphic state of character 59 has been gained
and lost six times, while the monophyly of clade 38 is supported by only a single state change in a character
(105) that has a very low consistency index (0-111). For this reason, I performed one final analysis of the
data set and will base my classification of the Plusiinae on this, rather than on the preferred 648-step
cladogram.

Analytical procedure

The data set was modified for re-analysis as follows. I considered the monophyly of the Plusiinae to have
been reasonably well-demonstrated and thus decided to remove the 11 outgroup taxa from the matrix. This
also allowed me to exclude 55 characters that exhibited state changes only among the outgroups. These
were: 8, 13, 17, 21, 23, 25, 26, 27, 30, 34, 36, 37, 38, 40, 43, 44, 45, 48, 53, 54, 55, 56, 57, 62, 66, 67, 68, 69,
70, 71, 72,:74, 75, 81, 82, 96, 115, 122, 123, 129, 135, IS1, 152, 153, 156, 161, 179, 180, it, 190, 195.
208, 210 and 211. A further five characters (20, 22, 46, 47, 114) were discarded because, after removal of
the outgroups, they became autapomorphic for a terminal taxon within the Plusiinae. Finally, all
characters that showed consistency indices of < 0-250 were deleted from the data set (5, 6, 12, 16, 18, 19,
42, 58, 59, 105, 107, 110, 144, 162, 191, 196, 201, 205, 209), with the following additions and exceptions.
Characters 65 (c = 0-750) and 83 (c = 0-313) were also removed. There were doubts regarding the
distinction of the states of the first; while the second (male sternal hair pencil development) showed
substantial homoplasy, and its relatively large consistency index is due, in part, to the large number of
states (5) scored. Ten characters were also left in the matrix despite having consistency indices of < 0-250:
3, 41, 97, 108, 131, 141, 148, 149, 189, and 214. These were generally characters that exhibited homoplasy |
only within the Plusiinae. Their apomorphic conditions also helped define high level cladesin the preferred |
648-step cladogram and might, therefore, be expected to exert a significant influence in a re-analysis. For
example, 141 appears to be potentially important in the Autoplusiina, while 189 concerns the possible
relationship between Syngrapha (s.1.), Anagrapha and clade 30.

As before, Shensiplusia was also omitted from the analysis.

Finally, the removal of the outgroups allowed character 14 to be analysed as an ordered character; all
other unordered characters were treated as before. i

The analysis was performed using the PAUP options HOLD = 3, SWAP = GLOBAL, MULPARS, |

NORELEASE; all other options being set to their default values. It was hoped that the deletion of about |

one-fifth of the taxa and over a third of the characters would speed up the computations. However, the
opposite result was obtained; the analysis slowed considerably and took an almost excessive amount of

computing time. I would therefore suggest that if anyone intends to delete such a large amount of data ‘in |_

program’, as was done here, then they circumvent the tree-building routine, which took most of the time,
by entering a USER-TREE that is expected to approximate the end result and perform branch-swapping
on that topology. An even better approach would probably be to recast the data physically to remove the

required taxa and characters and analyse this smaller data matrix. Either of these techniques should |~

decrease the run times of large reduced data sets significantly.

Results

As might have been expected, the MAXTREE default maximum of 50 cladograms was obtained. They |—
were 301 steps in length and had consistency indices of 0-584. Two observations can be made from these |
figures. First, exclusion of the outgroups and characters of low consistency eliminated over half the steps of |
the earlier cladograms (i.e. 346 of the original 647). However, the consistency index did not rise by as much |

as perhaps might have been expected, 0-475 to 0-584. This is primarily because the rise in consistency due to

the loss of the low consistency characters was counterbalanced to some degree by a lowering in consistency

due to the loss of many high consistency characters concerned with the outgroup interrelationships.

Examination of the 50 cladograms revealed a series of alternative subtopologies similar, but not
identical, to those obtained for the 648-step cladograms. (The clade numbers used subsequently are those
employed on the preferred cladogram and imply the same constituent taxa, though not necessarily the
same topology.)

First, the positions of Agrapha and Plusiopalpa varied. Three alternative topologies were recorded. On

_ the first, Plusiopalpa is placed as the sister-group of Argyrogramma + Stigmoplusia, with Agrapha as the
sister-group of Ctenoplusia + Acanthoplusia. The other two topologies are the same as those found for
clade 14 on the 648-step cladograms (see above).

Dactyloplusia has moved to become the sister-group of Thysanoplusia, rather than, as before, forming
the sister-group of Chrysodeixis + Pseudoplusia. This new clade is placed either as the sister-group of
Trichoplusia or as the sister-group of Trichoplusia + clade 12. Otherwise, the Argyrogrammatini display
the same topology as the preferred 648-state cladogram.

In the Euchalciina, there is ambiguity as to the position of Panchrysia + Chrysanympha. One
subtopology places this clade as the sister-group of Euchalcia + Adeva, while the other treats it as the
sister-group of clade 24, as on the preferred cladogram. Otherwise the Euchalciina is unchanged.

The exclusion of the characters of low consistency, as might have been expected, had a significant effect
upon the topology of the Autoplusiina and Plusiina. Clade 30 is recovered by and large intact, although
Loboplusia is placed as the sister-group of clade 32 + 35, with Erythroplusia as the sister-group to this
clade. However, Loboplusia and Erythroplusia can exchange positions.

At the base of the Autoplusiina, and within the Plusiina, there is what can only be described as near
chaos, with at least 13 alternative subtopologies being found. Syngrapha (s.1.) is recovered intact as before.
A new clade is also consistently obtained, comprising two pairs of taxa in a sister-group relationship;
Lophoplusia + Plusia and Autographa bractea + Cornutiplusia. However, the relationships of these,
Autographa gamma, clade 30, Anagrapha, Rachiplusia and Diachrysia appear to be able to take on any

| form. For example, at one extreme, Anagrapha forms the sister-group of Rachiplusia, which together form
the sister-group of the rest of the Autoplusiina, while Diachrysia forms the sister-group of clade 39. At the
other extreme, clade 37 is recovered intact but Rachiplusia is interpreted as the sister-group of clade 30 +
37. Numerous other alternative subtopologies were found, most of which comprise variations on the
placements of Anagrapha, Rachiplusia and Diachrysia. Such a result was not unexpected in view of the
evaluation of these taxa performed above and the characters deleted in the present analysis.

Along the main stem of the cladograms, no change in the topology was expected nor found. The cladistic
relationships among the tribes and subtribes is identical to that of the preferred 648-step cladogram,
namely: (Omorphinini (Abrostolini (Euchalciina (Autoplusiina + Plusiina)))).

To extract the maximally predictive classification of the Plusiinae from these alternatives, a manual strict
(Nelson) consensus cladogram analysis (Nelson, 1979) was performed. The ambiguity in clade 14 is
reduced to a terminal tetrachotomy between Plusiopalpa, Agrapha, Argyrogramma + Stigmoplusia and
Ctenoplusia + Acanthoplusia. Similarly, trichotomous relationships are assumed between clade 12,
Trichoplusia and Thysanoplusia + Dactyloplusia; and between clades 23, 25 and 26 in the Euchalciina.
Rachiplusia is removed from the Autoplusiina and placed with Anagrapha and Diachrysia in a pentachoto-
mous relationship with clades 30 and 37. This best reflects the uncertainty of relationship of the three single
genera concerned. Within clade 30, Loboplusia forms a trichotomy with Erythroplusia and clade 32 + 35;
while in clade 37, Autographa gamma, Syngrapha (s.\.) and Lophoplusia/Plusia/Autographa bractea/
Cornutiplusia are also trichotomously related.

Finally, in recognition of the fact that, on the preferred 648-step cladogram, neither clade 12 nor clade 14
was supported by good synapomorphies, and that, for example, blade-like valve setae (133) may
eventually prove to be a good synapomorphy for clades 13 and 16, the branches subtending clades 12 and 14
were collapsed, leaving a terminal heptachotomy within the Argyrogrammatini. This consensus cladogram
(with the addition of Pseudochalcia, Desertoplusia and Shensiplusia; see below) is shown in Figs 29, 30, and
will form the basis of the formal higher classification of the Plusiinae presented in this study.

| SYSTEMATICS OF THE PLUSIINAE F/I
;

The positions of Shensiplusia, Desertoplusia and Pseudochalcia

| Shensiplusia. Shensiplusia was described by Chou & Lu (1974) to accommodate the species S. nigribursa,
known only from the holotype female. It was said to resemble Argyrogramma (in the broadest sense) and,
specifically, to be very similar in external appearance to Acanthoplusia (as Argyrogramma) agnata. S.
nigribursa differed in being larger and in having more metallic iridescent scaling on the forewings. This

172 IAN J. KITCHING

PLUSIINI

Argyrogramma

Stigmoplusia

Plusiopalpa

Ctenoplusia

Acanthoplusia

Agrapha

Plusiotricha (Eutheiaplusia)
Plusiotricha (Plusiotricha)
Trichoplusia

Thysanoplusia
Dactyloplusia

Zonoplusia

Chrysodeixis (Chrysodeixis)
Chrysodeixis (Pseudoplusia)
Anadevidia

Shensiplusia

Abrostola

Mouralia

Omorphina

Fig. 29 The consensus classification of the tribes Omorphinini, Abrostolini and Argyrogrammatini. The
genus Shensiplusia is incertae sedis in the Argyrogrammatini.

description, together with Chou & Lu’s record for discovering ‘new’ species in China, suggests that S.
nigribursa may eventually prove synonymous with A. agnata or a related species. However, the female
genitalia of A. agnata figured by Ichinosé (19625: pl. 9, fig. 6) do not correspond exactly to those of S.
nigribursa (Chou & Lu, 1974: fig. 16). In particular, A. agnata does not appear to have the corpus bursae

covered internally with the numerous sclerotized spinules that give S. nigribursa its name. Chou & Lu

(1974) also stated that the corpus bursae of S. nigribursa was somewhat similar to that of Chrysodeixis (as
Neoplusia) acuta.

Although Chou & Lu provided little information useful in placing Shensiplusia within the Plusiinae, it
seems likely that the genus is an argyrogrammatine. However, further resolution of its position is
impossible at present (in particular, in the absence of the male) and in recognition of this, I have placed
Shensiplusia at the base of the Argyrogrammatini, in a trichotomy with Anadevidia and clade 6 (Fig. 29).

Needless to say, nothing is known of the biology of Shensiplusia, except that it was taken at Wukungin |

Shensi Province, China.

Desertoplusia. I learned of the descriptions of Desertoplusia and Pseudochalcia (Kljutshko, 1984) too late

to include their type species, D. bella (Christoph) and P. inconspicua (Graeser), in the full analysis. I thus |

have to be content with placing them as best I can onto the cladogram (Fig. 30) constructed from the other
genera. To this end, they will also provide useful tests of the robustness of the arrangement and its
supporting characters.

Desertoplusia contains two species, D. bella (Fig. 70) and D. paghmana (Wiltshire). In superficial
appearance, they resemble miniature autographas and were placed in that genus by Wiltshire (1971) and as
its sister-group by Kljutshko (1985b). However, an examination of the male genitalia of both species, as

SYSTEMATICS OF THE PLUSIINAE 173

Syngrapha (Caloplusia)

Syngrapha (Syngrapha)

Syngrapha (Palaeographa)
Lophoplusia

Plusia
Cornutiplusia
"Autographa" bractea
Autographa gamma
Anagrapha
Diachrysia
Rachiplusia
Autoplusia

Antoculeora

Sclerogenia

| Allagrapha
Macdunnoughia (Macdunnoughia)
Macdunnoughia (Puriplusia)
Loboplusia

| Erythroplusia

| Lamprotes

| Pseudeva
Plusidia

| Desertoplusia

| Euchalcia (Euchalcia)
Euchalcia (Adeva)
Panchrysia

| Chrysanympha

| Polychrysia

Eosphoropteryx

Pseudochalcia

Fig. 30 The consensus classification of the tribe Plusiini, comprising the subtribes Euchalciina, Autoplu-
siina and Plusiina. The genus Pseudochalcia is incertae sedis in the Euchalciina, while the genera
Rachiplusia, Diachrysia and Anagrapha are incertae sedis in the Autoplusiina + Plusiina. Loboplusia
and Erythroplusia are incertae sedis in the Autoplusiina.

174 IAN J. KITCHING

figured by Wiltshire (1971), reveals that they have simple oval valves (147), reduced clavi (144:1) and
subapical cornuti that arise from circular, concave plates (164). These character states indicate that
Desertoplusia belongs respectively to clades 21, 24 and 25. However, Desertoplusia retains the plesiomor-
phic plusiine forewing colour pattern and thus in Fig. 30, it is placed as the sister-group of Euchalcia +
Adeva. However, it should be noted that future studies may position the genus internal to clade 25.

D. bella is recorded in the BMNH collection only from Ashkhabad in Turkmeniya S.S.R. D. paghmana
comprises two subspecies. D. p. paghmana is confined to the Paghman Mountains, north-west of Kabul in
Afghanistan, while D. p. paupera is found only near Quetta, Pakistan (Wiltshire, 1971). Nothing is known
about the biology of either species but they appear to be part of the same steppe-montane fauna as
Euchalcia and Pseudochalcia. Being a euchalciine, the larval foodplant is predicted to be a member of the
Ranunculaceae or, much less likely, the Boraginaceae.

Pseudochalcia. I mentioned above in the clade analysis that Kljutshko (1984) had erected Pseudochalcia to
accommodate P. inconspicua (Fig. 69), thereby formalizing Dufay’s (1968) conclusion that the species did
not belong in Euchalcia. Kljutshko (1985b) interpreted Pseudochalcia to be the sister-group of Euchalcia,
based on the shape of the juxta apex. In both genera, this forms a somewhat recurved, blunt lobe. In this
study, however, a similar condition was found in Diachrysia chryson, Autographa gamma, Erythroplusia
rutulifrons, Antoculeora ornatissima, Adeva albavitta and Chrysanympha formosa. Examination of the
male and female genitalia figured by Kostrowicki (1961) yields little further data to help place Pseudochal-
cia. The valve is roughly quadrate in shape and thus resembles certain Euchalcia species. However, the
subapical cornutus is absent, the vesica ornamentation consisting primarily of a transverse row of straight
cornuti about halfway(?) along the vesica. The sacculus is armed apically with two short, divergent, sharp
points (saccular sheets?) and appears to have a basal, broad triangular point. These features suggest a
possible relationship with certain autoplusiine taxa. However, the apex of the corpus bursae is membra-
nous and lacks sclerotized ridges. Thus, like Dufay (1968), I do not know what the detailed relationships of
Pseudochalcia are. It is certainly a member of the Plusiini and possibly the Euchalciina, although inclusion
in the Autoplusiina cannot be entirely ruled out. In recognition of this uncertainty, I have placed
see in a trichotomous relationship in the Euchalciina with Eosphoropteryx and clade 21 (Fig.
30).

Nothing is known of the biology of P. inconspicua. It is restricted to the mountains of Central Asia
(Kostrowicki, 1961; Kljutshko, 1984).

The higher classification of the Plusiinae

The formal classification of the plusiine generic and higher level taxa given in Appendix 5 was constructed
using the conventions for annotated Linnaean hierarchies defined and described by Wiley (1979; 1981).
The hierarchy is fully sequenced (Nelson, 1974), that is, ‘taxa forming an asymmetrical part of a phylogenetic
tree may be placed at the same categorical rank and sequenced in phylogenetic order of origin’ (Wiley,
1981: 209). This has the effect of reducing the number of named subordinate taxa, the proliferation of
which has long been a target for anti-cladists (e.g. Colless, 1977). I have reduced this number further by
adopting informal genus-groups and, twice, genus-complexes. This approach was adopted because of the
intrinsic uncertainty regarding several areas of the underlying cladogram.

Taxa forming parts of polytomies ‘are given equivalent ranks and placed sedis mutabilis at the level of the
hierarchy at which their relationships to other taxa are known’ (Wiley, 1981: 211). This convention
differentiates those parts of the sequences that are dichotomous from those that are polytomous. It
indicates that the taxa labelled sedis mutabilis are of interchangeable position.

Finally, five genera were identified as being incertae sedis, of uncertain position. They are ‘placed in the
hierarchy ... at the level their relationships are best understood’ (Wiley, 1981: 212). For example,
Shensiplusia is considered to be an argyrogrammatine, but its sister-group could be the whole tribe or any
clade within the tribe. It is thus placed incertae sedis at the level of the Argyrogrammatini.

The application of these conventions allows the construction of a Linnaean hierarchy from which the
structure of the underlying cladogram is immediately apparent. Further conventions were formulated by
Wiley (1979; 1981) but were not necessary in the present context.

|
|

SYSTEMATICS OF THE PLUSIINAE 1S)

The Plusiinae are divided into four tribes: the monotypic Omorphinini Chou & Lu, the Abrostolini
Eichlin & Cunningham, the Argyrogrammatini Eichlin & Cunningham and the Plusiini Boisduval. The
Abrostolini comprises two genera, Abrostola and Mouralia.

Within the Argyrogrammatini, as noted above, Shensiplusia is placed incertae sedis. The first level of
sequencing comprises three genera (Anadevidia, Chrysodeixis and Zonoplusia) and the terminal Argyrog-
ramma-complex of genera and genus-groups. Chrysodeixis contains two subgenera, Chrysodeixis (s.str.)
and Pseudoplusia stat. n. The Argyrogramma-complex consists of seven lineages, all placed sedis
mutabilis. These consist of five genera (Trichoplusia, Plusiotricha, Ctenoplusia, Agrapha and Plusiopalpa)
and two genus-groups. Genus-groups were erected generally where the evidence in favour of synonymy
was considered inadequate. Thus Thysanoplusia stat. rey. and Dactyloplusia are placed together in the
Thysanoplusia genus-group, and Argyrogramma and Stigmoplusia in the Argyrogramma genus-group. In
contrast, Eutheiaplusia stat. n. and Acanthoplusia stat. rey. are reduced to subgeneric rank within
Plusiotricha and Ctenoplusia respectively. This approach, rather than full synonymy, was adopted because
I consider that both Eutheiaplusia and Acanthoplusia represent distinct monophyletic units within their
respective genera (see Clade analysis). As concluded above (Clade analysis), and contrary to the
(unstated) opinions of Franclemont & Todd (1983), I have found no convincing evidence for accepting a
synonymy of Ctenoplusia with Agrapha. I thus reinstate Crenoplusia stat. rev. as a genus separate from
Agrapha.

The Plusiini is divided into three subtribes: the Euchalciina Chou & Lu, the Autoplusiina subtrib. n. and
the Plusiina Boisduval.

Within the Euchalciina, Pseudochalcia is placed basally incertae sedis. The first level of sequencing
comprises Eosphoropteryx, Polychrysia and the Panchrysia, Euchalcia and Lamprotes genus-groups. The
last three of these are placed sedis mutabilis. They contain, respectively, Panchrysia and Chrysanympha,
Desertoplusia and Euchalcia, and (sequenced) Plusidia, Lamprotes and Pseudeva. Following the sugges-
tions of Kostrowicki (1961), Dufay, (1968) and Franclemont & Todd (1983), Adeva is placed within
Euchalcia, but as the subgenus Euchalcia (Adeva) stat. n., until such time as the internal structure of
Euchalcia (s.str.) is resolved.

Although the taxon Autoplusiina + Plusiina is not formally named, three genera are placed incertae sedis
at this level in the hierarchy. Rachiplusia will probably be shown eventually to belong to the Autoplusiina
but I can only speculate that this may also be the correct tribe for Anagrapha and Diachrysia.

Within the Autoplusiina, Erythroplusia and Loboplusia are placed basally incertae sedis. The remaining
two genus-groups form a fully dichotomous and sequenced clade. The Autoplusia genus-group comprises
Sclerogenia, Antoculeora and Autoplusia; the Macdunnoughia genus-group consists of Allagrapha and
Macdunnoughia. The latter genus includes Puriplusia as a subgenus.

Three groups are recognized within the Plusiina, in trichotomous relationship. That represented by
Autographa gamma is probably paraphyletic and requires further work. Syngrapha (s.1.) is recognized as a
monophyletic group comprising three sequenced subgenera (at least): Palaeographa, Syngrapha (s.str.)
and Caloplusia stat. rev. The final taxon in this group is the Plusia-complex, comprising two genus-groups
based upon Plusia and Cornutiplusia stat. rey. The former also includes Lophoplusia stat. n., the latter, the
group of Autographa represented by A. bractea. Given that there is at least one further lineage within that
genus (those species possessing seta SV2 on segment A1 of the larva), Autographa is at least triphyletic and
ought eventually to be subdivided. Until a thorough investigation is undertaken, however, formal naming
of the various lineages should be avoided.

Comparison with previous classifications and phylogenies

After Eichlin & Cunningham (1978) proposed their phylogeny of the Nearctic Plusiinae, they noted that
although it and those of Kostrowicki (1961) and Ichinosé (1962b) had been ‘derived independently . . .
Studying different faunal areas . . . the basic evolutionary lines are essentially the same’. Subsequently,

‘two further dendrogrammatic representations of the higher classification of the Plusiinae were published

(Chou & Lu, 1979c; Kljutshko, 1985a; b). A comparison of these five hierarchies with that proposed in the
present study shows that Eichlin & Cunningham’s observation still, by and large, holds true. However,
there are discrepancies between each and my own conclusions that deserve brief discussion.

Kostrowicki (1961) (Fig. 1 v Figs 29, 30). Kostrowicki’s study was, by a narrow margin, the first new
attempt to analyse the higher classification of the Plusiinae since Hampson (1913). It was based upon the
Palaearctic fauna and thus included very few representatives of the primarily tropical Argyrogrammatini.

176 IAN J. KITCHING

If it is assumed that Kostrowicki would have converted his ‘scheme of relationships’ directly into a
classification (which is unlikely but a necessary assumption if a meaningful comparison is to be made), then
the following points of concordance and disagreement with my classification can be noted.

Abrostola and Mouralia were considered to be the most primitive group of plusiines (stated implicitly).
Omorphina was erroneously transferred, without explanation, to the Acontiinae (as Erastriinae). Within
the Plusiinae examined, Kostrowicki’s first split was into two groups roughly corresponding to the
Autoplusiina + Plusiina and the Euchalciina + Argyrogrammatini. Misidentifications allowed him to
misplace Macdunnoughia in the Argyrogrammatini, while the other two autoplusiine genera considered,
Rachiplusia and Autoplusia, were treated together as the closest relatives of Cornutiplusia. Chrysanympha
was mistakenly associated with Plusia, from which it was supposed to have been derived. The Plusiina was
thus paraphyletic. The topology of the genera of this subtribe was similar to that in Fig. 30 but both Plusia
and Cornutiplusia branched off before Autographa, and Anagrapha was associated with Syngrapha (s.1.).

Within. the Euchalciina (which included Diachrysia), a similar topology to that in Fig. 30 was found.
However, the branching sequence for the Lamprotes genus-group was inverted and Panchrysia was placed
with Polychrysia.

The Argyrogrammatini, with the exception of Anadevidia, was monophyletic (allowing for Macdun-
noughia, see above), but Chrysodeixis was associated with Argyrogramma rather than with Pseudoplusia.
Anadevidia was treated as the sister-group of the rest of the Argyrogrammatini + Euchalciina.

Overall, most of the tribes and subtribes (with the exception of the Autoplusiina) are represented
reasonably accurately on Kostrowicki’s phylogenetic scheme. However, their interrelationships and
several of the generic assignments are at considerable variance with subsequent classifications.

Ichinosé (19625) (Fig. 2 v Figs 29, 30). Following very shortly after that of Kostrowicki (1961), Ichinosé’s
(19625) scheme of Japanese plusiine systematics represented a significant improvement due primarily to
the inclusion of Abrostola in the study. The recognition of this genus as the most ‘primitive’ plusiine taxon
examined resulted in Ichinosé reconstructing the first few branch-points within the subfamily more
accurately than Kostrowicki. Nevertheless, the Argyrogrammatini were paraphyletic, with Chrysodeixis
(sensu Trichoplusia (s.str.), Chrysodeixis (s.str.) and Ctenoplusia (s.1.)) branching off first, followed by a
pentachotomy consisting of Ctenoplusia (Acanthoplusia) (as Chrysodeixis) agnata, Thysanoplusia +
Anadevidia, Diachrysia, the Euchalciina and the Autoplusiina + Plusiina. As for Kostrowicki, this
interpretation of Ichinosé’s diagram requires its conversion into cladogram form. The original consisted of
a series of interconnected circles and thus had numerous extant taxa as putative ancestral forms to other
extant taxa. Hence some of the branches are zero-length projections (e.g. the branch subtending
Ctenoplusia (Acanthoplusia) agnata). The Euchalciina and the Plusiina were less well-resolved topologies
of those found in the present study, while the Autoplusiina (represented by Erythroplusia and Macdun-
noughia (s.1.) (including Sclerogenia and Antoculeora) was paraphyletic.

Thus Ichinosé’s scheme shows a considerable degree of concordance with that proposed here. The
discrepancies can be largely attributed to his inclusion of only two character complexes in the construction
of his systematic arrangement — larval chaetotactic ‘plan’ and the presence/absence of abdominal
dorso-lateral hair tufts in the males. This accounts for why Chrysodeixis (sensu Ichinosé) was biphyletic
and the Plusiini were not recognized as a monophyletic group (‘C.’ agnata lacks vestigial larval prolegs on
segments A3 and A4), and why the Autoplusiina were paraphyletic (both Macdunnoughia [sensu
Ichinosé’s ‘Scleroplusia’| and Autographa are heterogeneous regarding the presence/absence of seta SV2
on segment A1). In fact, Ichinosé’s interconnected circles are probably more correctly interpreted as
depicting the supposed evolution of plusiine larval chaetotaxy than of the Plusiinae themselves.

Eichlin & Cunningham (1978) (Fig. 3 v Figs 29, 30). Eichlin & Cunningham’s (1978) phylogeny of the
plusiines is by far the best published approximation to my own results. It incorporated both larval and adult
characters, with little apparent bias towards either, although the explicit use of some characters was often
unstated. The result was remarkably congruent with my own. All the tribal and subtribal groups confirmed
in this study were first recognized by Eichlin & Cunningham, with the exception of the Euchalciina and the
clade Autoplusiina + Plusiina.

Thus, the Abrostolini were placed as the most primitive tribe, followed by the basic subdivision of the
plusias into the Argyrogrammatini and Plusiini. Further analysis of the former was hindered by their virtual
absence from the Nearctic fauna, only five species from three genera occurring north of Mexico. The
Autoplusiina were the first to branch off the Plusiini lineage. Rachiplusia and Diachrysia were interpreted
as the sister-groups of Autoplusia and Allagrapha respectively, conclusions that are consistent with, if not
entirely supported by, my own results. The branching sequence of Polychrysia, Adeva and Pseudeva was
identical (allowing for missing Palaearctic taxa) but Chrysanympha and Eosphoropteryx were misplaced as

|
|
|

|

SYSTEMATICS OF THE PLUSIINAE 177

possible branches from the Plusiina. However, Eichlin & Cunningham were unsure of the position of these
enera.

F The non-monophyletic nature of Autographa, confirmed in the present study (if not in the manner
originally intended!), was first explicitly noted by Eichlin & Cunningham, although it must be said that this
conclusion was implicit in Ichinosé’s (1962b) scheme. Eichlin & Cunningham’s (1978) arrangement of
Plusia and Syngrapha (s.1.) agrees more or less with my own, as far as the two are comparable, with the
exception of their positioning of Anagrapha within Syngrapha. Their conclusions might have been
different had Macdunnoughia occurred in North America.

Chou & Lu (1979c) (Fig. 4 v Figs 29, 30). The phylogeny of the Plusiinae proposed by Chou & Lu (1979c)
represented a retrograde step in the systematics of the Plusiinae. It was overtly phenetic in derivation, very
poorly resolved and based upon relatively few characters, several of which were of very dubious quality and
reliability. However, they did recognize Omorphina (Omorphinini) as the most basal plusiine lineage,
followed by the Abrostolini (Abrostola). If their ‘Plusiini’ (Diachrysia, Anadevidia, Thysanoplusia) is
ignored, then the Argyrogrammatini is recovered intact, though it also includes Erythroplusia and
Puriplusia. The Euchalciina is also regarded as monophyletic. However, the internal relationships of these
two groups bear only minimal correspondence with my results. Syngrapha (s.1.) was interpreted as a
lineage independent of the Plusiina + Autoplusiina (less Puriplusia and Erythroplusia). Thus, despite
inadequate methods and reliance upon a few tried-and-tested ‘classic’ characters, several of the most
robust plusiine clades did emerge intact from Chou & Lu’s analysis.

_Kijutshko (1985a, b) (Figs 5-7 v Figs 29, 30). I described in detail above why Kljutshko’s original

cladogram (Fig. 5) of the Palaearctic Plusiinae was erroneous. After correctly applying cladistic methodol-
ogy, the somewhat more polytomous cladogram (Fig. 6) is comparable with my own findings. Like most
workers, Kljutshko recovered the Argyrogrammatini (less Anadevidia) as a monophyletic unit. The
Plusiini were also treated as monophyletic, allowing for the exclusion of Erythroplusia. However, the
fragmentation of the Euchalciina into four separate lineages resulted in almost complete incongruence
between Kljutshko’s groupings and my own. The only concordant taxa are Syngrapha + Caloplusia and the

_ Autoplusiina (represented by Macdunnoughia and Antoculeora). Reanalysis of some of his characters

(Fig. 7) resulted in the gain of only one more congruent clade, the Argyrogrammatini + Plusiini. Overall, a
laudable effort at explicit cladistic character analysis but a poor result.

Discussion
Larval foodplant associations

One of the most popular eco-systematic subjects at present is the investigation of the interac-
tions between herbivorous insects and their foodplants. The plusiines show a marked divergence
in the number and relationships of the plants that their larvae eat. Of those that are known (a
proviso that underlies this entire section of the discussion), many plusiine genera are extremely
polyphagous, often giving the impression that a plant merely has to be green, and preferably
herbaceous, to be a potential meal. Such genera include Chrysodeixis, Trichoplusia, Thysano-
plusia, Argyrogramma, Allagrapha, Autoplusia, Anagrapha and Autographa, which include
most of the notorious agricultural pests of the subfamily.

At the other extreme, a number of genera are known to be virtually monophagous at the level
of plant genus. For example, Lophoplusia violacea is known to eat only the labiate Stenogyne,
while its congener, L. pterylota, is recorded only from Clermontia in the relatively unrelated
Lobeliaceae. Many other examples were noted above and can be summarized thus: Abrostola
(Urticaceae and related families, one species on Asclepiadaceae); Mouralia (Commelinaceae);
Anadevidia (Cucurbitaceae); Ctenoplusia albostriata + putative allies (Asteraceae); Euchal-
ciina (most) (Ranunculaceae); Euchalcia (part) (Boraginaceae); Chrysanympha (Ericaceae);
Erythroplusia, Macdunnoughia, Sclerogenia, Antoculeora, Autographa bractea + putative allies
= Asteraceae); Plusia (monocotyledons); and Syngrapha celsa and rectangula species-groups
conifers).

The most striking observation about this wide ranging monophagy and oligiphagy is that there
is very little pattern to it from the point of view of either plant or plusiine systematics. The
sister-group of an oligophagous plusiine genus is about as likely to be polyphagous as oligopha-

178 IAN J. KITCHING

gous, and if the latter then it will probably feed on a markedly different group of plants.
Compare, for example, the sister-group pairs of Plusia/Lophoplusia, Abrostola/Mouralia and
Autoplusia/Antoculeora. The only two generalizations that can be made are that most euchal-
ciines feed on Ranunculaceae, and oligophagous autoplusiines seem to show a preference for
Asteraceae. Many foodplant records are missing for the tropical argyrogrammatines. When
their life histories are elucidated, I expect equally varied, interesting and oligophagous patterns
to be found as are presently observed in the remainder of the subfamily.

Biogeography

Very little has been published on the biogeography of the Plusiinae, mainly because all of the
previous studies have examined isolated faunas. This makes a world-wide comparison and
evaluation virtually impossible, especially from the historical viewpoint.

Kostrowicki (1961) presented in detail the descriptive and distributional biogeography of the
plusias, concentrating on the Palaearctic species. He noted that the Argyrogrammatini (his
group A) were essentially tropical, occurring in the Holarctic region primarily in the southern
areas. He considered his group B (Autoplusia, Rachiplusia, Cornutiplusia) to be characteristic
of Notogaea. Notogaea is defined by the Chambers Twentieth Century Dictionary as Australia,
the islands to the north, New Zealand and Polynesia, while the Concise Oxford English
Dictionary also includes the neotropics. Neither of these concepts coincides with the range of
group B, which is absent from Oceania but present also in Africa, and, to a lesser extent, in the
Holarctic. Kostrowicki’s group C (Syngrapha, Caloplusia, Anagrapha) was exclusively Holarc-
tic (with the exception of one unnamed species in the Andes — presumably Rachiplusia virgula,
which thus belongs to group B), as was group D (the balance of the plusias), although most
species of the latter group were Palaearctic.

Kostrowicki proceeded to examine these distributions in greater detail and speculated on the
centres of origin and dispersal routes of the various taxa. Group A was supposed to have
originated in the Old World tropics, while group C was of ‘obvious boreal-alpine and superficial-
ly circum-boreal origin’. Group D was hypothesised to have had its origins in the eastern
Palaearctic, where most of its species are currently found. However, lack of a consideration of
the Omorphinini and Abrostolini, coupled with a dispersalist/centre of origin approach resulted
in much speculative fiction. This was compounded by Kostrowicki’s errors of identification and
his non-critical systematic analysis.

A better basis for a zoogeographical analysis was Eichlin & Cunningham’s (1978) phylogeny.
However, their comments on the subject comprised merely a single paragraph. Coloured by
their belief that the Plusiinae were very closely related to the Stictopterinae, they postulated a
tropical origin for the plusiines, followed by an invasion of the temperate zones. The Argyro-
grammatini were again recognized as a primarily tropical tribe, the Plusiini as primarily
temperate, some of which had further adapted to Boreal or Arctic conditions.

The results of the present study represent the first attempt to analyse the higher classification
of the Plusiinae on a world-wide basis. It would thus probably be seen as unforgivable if I did not
make some statements regarding the historical biogeography of the subfamily. However, they
will necessarily be non-commital and somewhat conjectural, given the ambiguity contained in
the cladogram upon which they are based.

In the absence of a reasonable hypothesis as to the immediate outgroup to the Plusiinae, I can
only speculate where the subfamily might have first arisen. Perhaps the marginally most
plausible area is the south-eastern Palaearctic/north-eastern Oriental regions. In this area are
currently found the most plesiomorphic plusiine, Omorphina; the Abrostola species (A.
anophioides) that may represent the intermediate link with Mouralia; several of the more
plesiomorphic argyrogrammatine genera (notably Anadevidia and Zonoplusia); and some of
the plesiomorphic members of the three subtribes of the Plusiini: Diachrysia, Erythroplusia,
Polychrysia.

If this hypothesis is accepted, for the sake of argument, then the ‘proto-Abrostolini’ may have
had a more or less world-wide occurrence, with subsequent vicariance events isolating first

-
4
:
{
:
'

|

SYSTEMATICS OF THE PLUSIINAE 179

Mouralia, then the Nearctic abrostolas. A cladistic analysis of the Abrostolini is necessary to test
this suggestion.

It appears most likely that the Argyrogrammatini evolved primarily in the tropics and the
Plusiini in the temperate regions. Superficially, this may suggest a Gondwanan and Laurasian
distribution pattern respectively. However, it is extremely unlikely that the cladogenetic event
that ultimately gave rise to the two major plusiine lineages was contemporaneous with the
primary break-up of Pangea. Such an association would place the time of divergence of the
Argyrogrammatini and Plusiini at between 180 (Rosen, 1978) and 150 (Irving, 1977) million
years ago, that is, in the Jurassic, before the currently accepted origin of the angiosperms! A
more recent origin would seem less incredible but remains unknown.

There are other interesting patterns of distribution within the subfamily. Chrysodeixis seems
to show a good example of a transantarctic distribution. C. (Pseudoplusia) is found in South
America and Saint Helena, while the hypothesised plesiomorphic C. (Chrysodeixis), C.
argentifera, is found in Australia. More apomorphic Chrysodeixis species are then found
towards the Equator. This genus would be one of the most profitable to study from a
biogeographic point of view, as well as containing no fewer than five economically important
pest species (C. includens, C. argentifera, C. eriosoma, C. chalcites and C. acuta).

Within the Plusiini, there are several examples of Nearctic/Palaearctic sister-genus pairs:
Chrysanympha/Panchrysia, Euchalcia (Adeva)/Euchalcia (s.str.), Pseudeva/Lamprotes, Alla-
grapha/Macdunnoughia and Autoplusia/Antoculeora. These have only become apparent follow-
ing a world-wide analysis of the subfamily and all merit further investigation. The conclusion
that Lophoplusia is the sister-group of Plusia implies that the former, a Hawaiian endemic,
originated by dispersal from either the eastern Palaearctic or western Nearctic. A decision
between the two must await the resolution of the interrelationships of the species of these, and
maybe related, genera. Finally, the most apomorphic syngraphas appear to be those that live
furthest north, culminating in those with a circumpolar distribution. A cladistic analysis is
necessary to reveal whether this northward pattern is truly as linear as it first appears.

All of the above observations are preliminary and tentative, and are meant more as stimuli to
further research than as statements of fact. More concrete conclusions must await detailed
cladistic investigations of the individual plusiine genera and a more comprehensive analysis of
the higher classification. There is no doubt, however, that the Plusiinae could prove to be most
rewarding subjects for biogeographical analysis.

Conclusions

Eichlin & Cunningham (1978) concluded with the hope that their study would be useful as a
framework for future work. By examining the world fauna, I have endeavoured to synthesize an
overall classification that has illuminated transcontinental relationships not suspected from the
previous faunal studies. However, this arrangement is also preliminary, as any first attempt in an
uncharted field must be. Although I have referred in this paper to over half the plusiine species, I
have examined in detail less than one-eighth of them. Many species still languish in ‘“Plusia’ s.1.
and the elucidation of their relationships should have high priority, if only to empty a dustbin.
Similarly, Trichoplusia ought to be dismembered into its component monophyletic groups but I
Suspect that an analysis of the ‘higher’ argyrogrammatines may have to be performed first or
simultaneously, rendering this a massive task. The subtribes of the Plusiini are also poorly
defined and grey areas often include potential or actual pest species (e.g. Anagrapha falcifera).

Autographa, in particular, is in dire need of attention but would have to be analysed in

conjunction with the rest of the Plusiina.

There are numerous biological topics that could employ plusiines as subjects. Perhaps the
most important, as with all other groups of Lepidoptera, is the elucidation of life histories, to
open up the possibilities for employing early stages in systematic analyses. The plusiines also
lend themselves to questions of insect/plant interactions (How and why did Abrostola asclepiadis
switch from Urtica to Vincetoxicum? Why does one Lophoplusia feed on a scentless mint and
another on a tree lobelia with sticky, milky latex?); of pheromone biology (What does Mouralia

180 IAN J. KITCHING

do with its giant hairpencil? Why do Loboplusia and Chrysodeixis have such elaborate scent
organs?); and of biogeography (Chrysodeixis and Autoplusia are the prime candidates). If
nothing else, I hope that this study will stimulate interest in the group so that such problems will
be investigated. The Plusiinae could be much more than gold spots and burnished brasses,
spectacles and silver Ys.

Addenda: notes added in proof
Taxa examined

Brachionycha sphinx Hufnagel was inadvertently used as the exemplar species for this genus
instead of the type species, B. nubeculosa Esper.

Clade analysis

The larva of Abrostola triopus Hampson has been recorded feeding on Commelina sp. and
the grass, Pennisetum purpureum (Sevastopulo, 1976). The former record is of particular
interest as Commelina belongs to the same family (Commelinaceae) as the larval foodplants of
Mouralia, the sister-genus of Abrostola. A. triopus is found in South Africa (with three doubtful
records from Ethiopia and Kenya, probably misidentifications) and thus may be a critical species
in the future understanding of the evolution of the Abrostolini.

Sevastopulo (1976) also recorded the larva of Trichoplusia vittata (as transfixa) feeding on
Artemisia africana, so corroborating the hypothesised oligophagy of this group.

The larval foodplant of Plusiopalpa dichora (as P. adrasta) was given by Sevastopulo (1976) as
Coleus (Lamiaceae).

An undescribed species of Euchalcia with yellow hindwings has recently been discovered in
north-western Canada (J. D. Lafontaine, pers. comm.).

Polychrysia morigera appears to be extending its range eastwards in the U.S.A. (Covell &
Medley, 1986). It was recorded from Kentucky in 1976, from Pennsylvania (recent unknown
date) and was caught for the first time in Tennessee in 1984, trapped in the flowers of a Lady’s
Slipper Orchid (Cypripedium kentuckiense)!

Lafontaine (in press) has recently shown that there has been a considerable degree of
confusion regarding the correct identity of Syngrapha ottolenguii. The type is a female that has
had the abdomen of a male Autographa californica glued on. Examination of true ottolenguii has
shown the species to be closely related to S. (Palaeographa) interrogationis and S. (P.) nyiwonis.
Thus, S. ottolenguii almost certainly belongs in the subgenus Palaeographa of Syngrapha. In
addition, the series examined by Eichlin & Cunningham (1978) was apparently mixed. When
sorted correctly, S. ottolenguii was found to be endemic to Attu Island in the Aleutian Islands.
‘S. ottolenguii’ in Alaska and Canada was shown by Lafontaine (in press) to be Autographa
buraetica Staudinger, a species previously known only from the eastern Palaearctic and, until
recently, considered as a subspecies or even a form of Autographa pulchrina.

Lafontaine (in press) has also examined a large number of specimens of Syngrapha devergens
and S. alticola. He concluded that the genitalic differences noted by McDunnough (1944) are
constant and that the Nearctic S. alticola should be reinstated as a species separate from the
Palaearctic S. devergens.

References

Aistleitner, E. & Rezbanyai, L. 1982. Neu fiir Osterreich: Diachrysia nadeja Obth. 1880 auch nérdlich der
Alpen nachgewiesen (Noct.). Nota Lepidopterologica 5: 55-56.

Alphéraky, S. 1892. Lepidoptera nova a Gr. Grum-Grshimailo in Asia Centrali. Novissime lecta. Horae
Societatis Entomologicae Rossicae 26: 444—459.

Angulo, A. O. 1973. Estados preimaginales de Syngrapha gammoides (Blanchard), (Lepidoptera: Noc-
tuidae). Boletin de la Sociedad de Biologia de Concepcion 46: 155-162.

—— 1978. La subfamilie Phytometrinae en Chile (Lepidoptera: Noctuidae). Brenesia 14-15: 57-95.

Angulo, A. O. & Weigart, G. T. 1974. Rachiplusia nu (Guenée), estados inmaduros y biologia (Lepidop-
tera: Noctuidae). Boletin de la Sociedad de Biologia de Concepcion 48: 117-122.

SYSTEMATICS OF THE PLUSIINAE 181

_ — 1975. Estados inmurados de Lépidépteros Noéctuidos de importancia econdmica en Chile y claves

para su determinacién (Lepidoptera: Noctuidae). Sociedad Biologia de Concepcién. Publicacién
Especial 2: 1-153.

_ Barnes, W. & McDunnough, J. H. 1917. Checklist of the Lepidoptera of Boreal America. viii + 392 pp.

Decatur, Illinois.

Bayer, M. B. 1965. The comparative morphology and musculature of the male genital segments in six
species of the Plusiinae (Lepidoptera — Noctuidae). South African Journal of Agricultural Science 8:
171-182.

Beck, H. 1960. Die Larvalsystematik der Eulen (Noctuidae). 406 pp. Berlin.
_ Betala, S. R. & Kushwaha, K. S. 1980. Biology and external morphology of forage pests IV. Cabbage

semilooper, Plusia nigrisigna Walker on lucerne and cowpea. Bulletin of Entomology. Entomological
Society of India 13: 95-100.

Birch, M. C. 1972a. Male abdominal brush organs in British noctuid moths and their value as a taxonomic
character. Part I. Entomologist 105: 185-205.

— 1972b. Male abdominal brush organs and their value as a taxonomic character. Part II. Entomologist
105: 233-244.

Bogner, F., Boppré, M., Ernst, K.-D. & Boeckh, J. 1986. CO, sensitive receptors on labial palps of
Rhodogastria moths (Lepidoptera: Arctiidae): physiology, fine structure and central projection. Journal
of Comparative Physiology (A) 158: 741-749.

_ Boisduval, J. B. A. D. 1828. Europaeorum Lepidopterorum Index Methodicus . . . pars Pruna, sistens

Papilio, Sphinx, Bombyx et Noctua Linn. 103 pp. Paris.

Borkhausen, M. B. 1792. Naturgeschichte der europdischen Schmetterlinge 4, xvi + 809 pp. Frankfurt.

Bretherton, R. F., Goater, B. & Lorimer, R. I. 1983. Noctuidae (continued), pp. 36—414. Jn Heath, J. &
Emmet, A. M. (Eds), The moths and butterflies of Great Britain and Ireland. 10: Noctuidae (part II) and
Agaristidae. 459 pp. Colchester.

Butler, A. G. 1892. On the genus Polychrysia of Hiibner (a group of plusiid moths). Annals and Magazine
of Natural History; including Zoology, Botany and Geology (6) 9: 407—408.

Callahan, P. S. 1960. A morphological study of spermatophore placement and mating in the subfamily
Plusiinae (Noctuidae: Lepidoptera). Proceedings of the XI International Congress of Entomology:
339-345.

Canerday, T. D. & Arant, F. S. 1967. Biology of Pseudoplusia includens and notes on the biology
of Trichoplusia ni, Rachiplusia ou and Autographa biloba. Journal of Economic Entomology 60: 870-
871.

Carter, D. J. 1984. Pest Lepidoptera of Europe with special reference to the British Isles. 431 pp. Dordrecht.

Chao, Y. C., Young, III, S. Y. & Sim, K. S. 1984. Cytopathology of the soybean looper, Pseudoplusia
includens, infected with the Pseudoplusia includens icosahedral virus. Journal of Invertebrate Pathology
45: 16-23.

Chauvin, G. & Faucheux, M. 1981. Les piéces buccales et leurs recepteurs sensoriels chez l’imago de
Micropterix calthella (Lepidoptera: Micropterigidae). International Journal of Insect Morphology and
Embryology 10: 425-439.

Chou, I. & Lu, T. 1974. Studies on Chinese Plusiinae (Lepidoptera: Noctuidae). Acta Entomologica Sinica
17: 66-82.

_ — 1978. Studies on Chinese Plusiinae. Acta Entomologica Sinica 21: 63-78.

— 1979a. Eight new species of Plusiinae and revision of some of its known species (Lepidoptera:
Noctuidae). Acta Entomologica Sinica 22: 61-72.

— 1979b. Two new genera, four new species of Plusiinae and revision of some of its known species
(Lepidoptera: Noctuidae). Entomotaxonomia 1: 15-22.

— 1979c. A discussion on generic system of Plusiinae. Entomotaxonomia 1: 71-78.

Colless, D. H. 1977. A cornicuopia of categories. Systematic Zoology 26: 349-352.

Comstock, J. A. 1937. A new record and a life history (Lepidoptera). Bulletin of the Southern California
Academy of Sciences 35: 169-174.

| Covell, C. V. & Medley, M. E. 1986. Polychrysia morigera (Noctuidae) trapped in a Southern Lady’s

Slipper labellum in Tennessee. Journal of the Lepidopterists’ Society 40: 128-129.

Crumb, S. E. 1934. A classification of some noctuid larvae of the subfamily Hypeninae. Entomologica
Americana 14: 133-196.

— 1956. The larvae of the Phalaenidae. Technical Bulletin. United States Department of Agriculture no.
1135: 1-365.

Dahlgren, R. 1983. General aspects of angiosperm evolution and macrosystematics. Nordic Journal of
Botany 3: 119-149.

182 IAN J. KITCHING

[Denis, J. N. C. M. & Schiffermiiller, I.] 1775. Anktindung eines systematischen Werkes von den
Schmetterlingen der Wienergegend. 323 pp. Wien.

Dewick, A. J. 1951. Rarities in east Essex. Entomologist 63: 253.

Dierl, W. 1975. Loboplusia vanderweelei Roepke: a little known oriental noctuid (Lep.). Entomologische
Zeitschrift. Frankfurt a. M. 85: 245-246.

Draudt, M. 1937. Corrections and addenda to supplementary volume 3. 2. Family Noctuidae. In Seitz, A..,
The Macrolepidoptera Of the World. 3 (Suppl.): 1-333. Stuttgart.

— 1940. Subfamily: Phytometrinae. Jn Seitz, A., The Macrolepidoptera of the World. 7: 462—471.
(consulted in the German edition) Stuttgart.

Dufay, C. 1957. Les éspéces asiatiques du genre Abrostola O. (Unca auct.). (Lepidopt. Agrotidae).
Bonner Zoologische Beitrage 8: 307-321.

—— 1958. Descriptions de nouvelles Abrostola africaines (Lep. Noctuidae). Bulletin del’ Institute Francais
d’ Afrique Noire 20: 199-216.

1960. Les éspéces japonaises du genre Abrostola O. (Unca auct.) (Lepidoptera Noctuidae Plusiinae).

Tinea 5: 303-313.

1968. Revision des Plusiinae Paléarctiques I. Monographie du genre Euchalcia Hibner. Veréffent-

lichungen der Zoologischen Staatssammlung Miinchen 12: 21-154.

1969. Un Plusiinae nouveau pour la France: Chrysaspidia putnami (Grote) (= festata Graes., gracilis
Lempke, nova syn.) [Lep. Noctuidae]. Alexanor 6: 57-72.

— 1970a. Insectes Lépidoptéres Noctuidae Plusiinae. Fauna de Madagascar 31: 1-198.

—— 1970b. Descriptions de nouvelles éspeces et d’un genre de Plusiinae Indo-Australiens (Lep.
Noctuidae). Bulletin Mensuel de la Société Linnéene de Lyon 39: 101-107.

1971. Description d’Abrostola kaszabi n. sp. de Mongolie, et note synonymique sur des Abrostola

asiatiques (Lepidoptera, Noctuidae, Plusiinae). Reichenbachia 13: 269-273.

1972. Descriptions de nouveaux Plusiinae africains (Lep. Noctuidae). Bulletin Mensuel de la Société

Linnéene de Lyon 41: 65-72, 93-111.

1973. Les Plusiinae des expéditions allemandes au Népal de 1955 a 1967 (Lepidoptera, Noctuidae).

Ergebnisse Forschungsunternehmens Nepal Himalaya 4: 389-400.

1974. Descriptions de nouveaux Plusiinae Indo-Australiens et Neotropicaux (Lep. Noctuidae).
Bulletin Mensuel de la Société Linnéene de Lyon 43: 102-111.

—— 1975. Les Plusiinae de la Réunion. Bulletin de la Société Entomologique de France 80: 157—166.

—— 1977. Ergebnisse der Bhutan-Expedition 1972 des Naturhistorischen Museums in Basel. Lepidop-
tera: Fam. Noctuidae Subfam. Plusiinae. Entomologia Basiliensia 2: 139-143.

1982a. Description de nouveaux Plusiinae Indo-Australiens (Lépidopt. Noctuidae). Bulletin Mensuel

de la Société Linnéene de Lyon 51: 71-76.

1982b. Les Plusiinae de l’archipel des Comores. Bulletin de la Société Entomologique de France 87:
220-228.

—— 1986. Un Plusiinae nouveau pour la France: Diachrysia nadeja (Oberthiir), 1880 (Lepidoptera,
Noctuidae). Nota Lepidopterologica 9: 47—S0.

Duponchel, P. A. J. 1826. p.[3] Jn Godart, J.-B. & Duponchel, P. A. J. Histoire naturelle des Lépidopteres
ou Papillons de France. Nocturnes. Tome troisiéme. 6: 475 pp. Paris.

1829. p. 72 In Godart, J.-B. & Duponchel, P. A. J. Histoire naturelle des Lépidopteéres ou Papillons de

France. Tome quatriéme.- Deuxiéme Partie. 7: 507 pp. Paris.

1844. Catalogue méthodique des Lépidopteéres d’ Europe. xxx + 523 pp. Paris.

Dyar, H. G. 1896. Descriptions of two noctuid larvae. Journal of the New York Entomological Society 4:
145.

— 1902. A generic subdivision of the genus Plusia. Journal of the New York Entomological Society 10:
79-82.

Eichlin, T. D. 1975. Guide to the adult and larval Plusiinae of California (Lepidoptera: Noctuidae).
Occasional Papers in Entomology, California State Department of Food and Agriculture 21: 1-73.

Eichlin, T. D. & Cunningham, H. B. 1969. Characters for identification of some common plusiine
caterpillars of the southeastern United States. Annals of the Entomological Society of America 62:
507-510.

—— 1978. The Plusiinae (Lepidoptera: Noctuidae) of America north of Mexico, emphasizing genitalic and
larval morphology. Technical Bulletin. United States Department of Agriculture Agricultural Research
Service no. 1567: 1-222.

Esaki, T., Issiki, S., Mutuura, A., Inoue, H., Ogata, M., Okagaki, H. & Kuroko, H. 1971. Icones
heterocerorum Japonicorum in coloribus naturalibus. 2: 304 pp. Revised edition. Hoikusha, Osaka.

Farris, J. S. 1974. Formal definitions of paraphyly and polyphyly. Systematic Zoology 23: 548-554.

SYSTEMATICS OF THE PLUSIINAE 183

Fletcher, D. S. 1963. 1. Geometridae. 2. Noctuidae. Exploration du Parc National Albert (2) 15: 1-155.

Forbes, W. T. M. 1910. A structural study of some caterpillars. Annals of the Entomological Society of
America 3: 94-143.

—— 1914. A table of the genera of Noctuidae of northeastern North America. Journal of the New York
Entomological Society 22: 1-33.

1954. Lepidoptera of New York and neighboring states. Part 3. Noctuidae. Memoirs. Cornell
University Agricultural Experiment Station no. 329: 1—433.

Forster, W. & Wohlfahrt, T. A. 1971. Die Schmetterlinge Mitteleuropas 4. Eulen (Noctuidae). vii + 329 pp.
Stuttgart.

Fracker, S. B. 1915. The classification of lepidopterous larvae. Illinois Biological Monographs 2:
1-170.

Franclemont, J. G. 1964. Notes and two new genera (Lepidoptera: Noctuidae). Proceedings of the
Entomological Society of Washington 66: 71-73.

Franclemont, J. G. & Todd, E. L. 1983. Noctuidae, pp. 120-159. In Hodges, R. W. et al. Checklist of the
Lepidoptera of America north of Mexico. London.

Gaede, M. 1937-1939. Subfamily: Phytometrinae. Jn Seitz, A., The Macrolepidoptera of the World. 15:
256-263. Stuttgart.

Gardner, J. C. M. 1941. Immature stages of Indian Lepidoptera (2) [Noctuidae, Hypsidae]. /ndian Forest
Records (N.S.) 6: 253-296.

-—— 1947. On the larvae of the Noctuidae (Lepidoptera) — III. Transactions of the Royal Entomological

Society of London 98: 59-90.

-— 1948a. On larvae of the Noctuidae (Lepidoptera) — IV. Transactions of the Royal Entomological

Society of London 99: 291-318.

-— 1948b. Notes on the pupae of the Noctuidae. Proceedings of the Royal Entomological Society of

London (B) 17: 84-92.

Gerasimov, A. M. 1935. Zur Frage der Homodynamie der Borsten von Schmetterlingsraupen. Zoologis-
cher Anzeiger 112: 177-194.

Gothilf, S. & Shorey, H. H. 1976. Sex pheromones of Lepidoptera: examination of the role of male scent
brushes in courtship behavior of Trichoplusia ni. Environmental Entomology 5: 115-119.

Grant, G. G. 1971. Scent apparatus of the male cabbage looper, Trichoplusia ni. Annals of the
Entomological Society of America 64: 347-352.

Grote, A. R. 1874. List of the Noctuidae of North America. Bulletin of the Buffalo Society of Natural
Science 2: 1-77.

— 1875. Checklist of the Noctuidae of America north of Mexico. | Bombyciae and Noctuelitae
(Nonfasciatae). 28 pp. New York.

— 1882. New check list of North American moths. 73 pp. New York.

— 1895. The hypenoid moths and allied groups. Proceedings of the American Philosophical Society 34:
416-436.

— 1896. Generic types in Plusia. Entomologist’s Record and Journal of Variation 8: 303.

Guenée, A. 1841. Essai sur la classification des Noctuélides (suite) (1). Annales de la Société Entomologique

de France 10: 53-83.
— 1852. In Boisduval, J. A. & Guenée, A., Histoire naturelle des Insectes — Species général des
Lépidopteéres. 6 Noctuélites. I. 444 pp. Paris.

-Haggett, G. M. 1958. Larvae of the British Lepidoptera not figured in Buckler. Part III. Proceedings of the

South London Entomological and Natural History Society 1957: 94-99.

Hampson, G. F. 1893-1895. The fauna of British India, including Ceylon and Burma. 2, xxii + 609 pp.
London.

— 1902. The moths of South Africa (Part II). Annals of the South Africa Museum 2: 255-446.

— 1912. Catalogue of the Lepidoptera Phalaenae in the collection of the British Museum. 11 Eutelianae,
Stictopterinae, Sarrothripinae, Acontianae. xvii + 689 pp. London.

— 1913. Catalogue of the Lepidoptera Phalaenae in the collection of the British Museum. 13 Catocalinae

_ continued, Mominae, Phytometrinae. xvi + 609 pp. London.

Hardwick, D. F. 1970. A revision of the North American Heliothidinae (Lepidoptera: Noctuidae).
Memoirs of the Entomological Society of Canada 73: 1-59.

Haworth, A. H. 1809. Lepidoptera Britannica. Part Il: pp. 137-376. London.

Hayes, A. H. 1975. The larger moths of the Galapagos Islands (Geometroidea: Sphingoidea & Noc-
tuoidea). Proceedings of the California Academy of (Natural) Science 15: 145-208.

Herrich-Schaffer, G. A. W. 1845. Systematische Bearbeitung der Schmetterlinge von Europa, zugleich als
Text, Revision und Supplement zu Jakob Hiibner’s Sammlung europdischer Schmetterlinge. 2. Die

184 IAN J. KITCHING

Schwarmer, Spinner und Eulen (Hepialides-Cossides-Zygaenides-Sesiides-Sphingides-Bombycides-
Noctuides-Nycteolides). 450 pp. Regensburg.

Heywood, V. H. (ed.) 1978. Flowering plants of the World. 335 pp. Oxford.

Hinton, H. E. 1946. On the homology and nomenclature of the setae of lepidopterous larvae, with some
notes on the phylogeny of the Lepidoptera. Transactions of the Royal Entomological Society of London
97: 1-37.

Hirschke, H. 1898. Die ersten Stande von Plusia aemula Hb. Verhandlungen der Kaiserlich K6niglichen
Zoologisch-Botanischen Gesellschaft in Wien 1898: 1-2.

Hogue, C. L. 1963. A definition and classification of the tribe Stiriini (Lepidoptera: Noctuidae).
Contributions in Science. Los Angeles County Museum 64: 1-129.

Holloway, J. D. 1985. The moths of Borneo: family Noctuidae: subfamilies Euteliinae, Stictopterinae,
Plusiinae, Pantheinae. Malayan Nature Journal 38: 157-317.

Hoyningen-Heune, F. F. von 1901. Aberrationen einiger Estlandischer Eulen und Spanner. Berliner
Entomologische Zeitschrift 46: 309-319.

Hiibner, J. 1816—[1826]. Verzeichniss bekannter Schmettlinge 432 + 72 pp. Augsburg.

Hyslop, J. A. 1912. The alfalfa looper in the Pacific Northwest (Autographa californica Speyer). United
States Department of Agriculture Bureau of Entomology. Bulletin 95(7): 109-118.

Ichinosé, T. 1958a. Studies on the genus Plusia (Noctuidae) I. On the external features and identification of
the larvae of several species belonging to the genus Plusia. Part 1. [In Japanese with English summary. |
Kontyti 26: 86-97.

— 1958b. Studies on the genus Plusia (Noctuidae) II. On the external features and identification of the
larvae of several species belonging to the genus Plusia. Part 2. [In Japanese with English summary.]
Kontyu 26: 123-133.

—— 1959. Studies on the bionomics of the Asiatic common looper, Plusia nigrisigna Walker, and its
several allied species (Noctuidae). I. On the nomenclature of Plusia nigrisigna Walker. [In Japanese
with English summary. ] Japanese Journal of Applied Entomology and Zoology 3: 99-106.

1960a. Studies on the genus Plusia (Noctuidae) III. On the external features and identification of the

larvae of several species belonging to the genus Plusia. Part 3. [In Japanese with English summary. |

Kontyti 28: 1-8.

1960b. Studies on the bionomics of the Asiatic common looper Plusia nigrisigna Walker and its
several allied species (Noctuidae). III. Effects of temperature on the development and the pupal
colouration of the Asiatic common looper, Plusia nigrisigna Walker. [In Japanese with English
summary.]| Japanese Journal of Applied Entomology and Zoology 4: 26-30.

—— 1962a. Studies on the genus Plusia (s.1.) (Noctuidae, Plusiinae) IV. On Autographa group and
peponis group. Kontyii 30: 248-251.

—— 1962b. Studies on the noctuid subfamily Plusiinae of Japan. Bulletin of the Faculty of Agriculture of the
Tokyo University of Agriculture and Technology 6: 1-127.

— 1963a. The review of Kostrowicki (1961) on the palaearctic species of the subfamily Plusiinae. [In
Japanese.] Japan Heterocerists’ Journal 33: 248-249.

— 1963b. A new species belonging to the genus Syngrapha from Japan (Noctuidae, Plusiinae). Tinea 6:
59-62.

—— 1973. A revision of some genera of the Japanese Plusiinae, with descriptions of a new genus and two
new subgenera (Lepidoptera, Noctuidae). Kontyé 41: 135-140.

Ichinosé, T. & Asawa, T. 1964. Studies on the bionomics of the Asiatic common looper, Plusia nigrisigna
Walker and its several allied species (Lepidoptera, Noctuidae). IV. Effect of temperature on the pupai
colour of the Asiatic common looper. Japanese Journal of Applied Entomology and Zoology 8: 235-244.

Ichinosé, T., Hironobu T. & Ishino, Y. 1967. Hereditary nature of the two forms of Argyrogramma
albostriata (Bremer & Grey). (Noctuidae, Plusiinae). Kontyii 35: 329-336.

Ichinosé, T. & Shibuya, S. 1959. Studies on the bionomics of the Asiatic common looper, Plusia nigrisigna
Walker and its several allied species (Noctuidae). I. Effects of temperature on the development of the
Asiatic common looper, Plusia nigrisigna Walker. Japanese Journal of Applied Entomology and
Zoology 3: 157-163.

Inoue, H. & Sugi, S. 1958-61. Checklist of the Lepidoptera of Japan. 5. Noctuidae and 6 Hyblaeidae. pp.
431-683. Tokyo.

Irving, E. 1977. Drift of the major continental blocks since the Devonian. Nature 270: 304—309.

Izquierdo, V. S. 1895. Notas sobre los Lepidépteros de Chile. Anales de la Universidad de Chilé 40:
783-835.

Kasy, F. 1953. Phytometra (Plusia) zosimi Hbn.: tiber die ersten Stande, Biologie und Zuchtergebnisse.
Zeitschrift der Wiener Entomologischen Gesellschaft 38: 321-333.

SYSTEMATICS OF THE PLUSIINAE 185

Kearfott, W. D. 1904. Notes on the life-history of Polychrysia formosa Gr. Entomological News 15:
301-302.

Kimball, C. P. 1965. Arthropods of Florida and neighboring land areas. Lepidoptera of Florida, an
annotated check list. v. + 363 pp. Gainesville.

Kirby, W. F. 1889. European butterflies and moths. \vi + 427 pp. London.

1907. The butterflies and moths of Europe. \xxii + 432 pp. London.

Kiriakoff, S. G. 1970. Die Stellung von Diloba caeruleocephala (L.) im Lepidopteren-System. Nachrich-
tenblatt der Bayerischen Entomologen 19: 101-104.

Kitching, I. J. 1984. An historical review of the higher classification of the Noctuidae (Lepidoptera).
Bulletin of the British Museum (Natural History). (Entomology) 49: 153-234.

Kljutshko, Z. F. 1983. New data on taxonomy of the genus Syngrapha (Lepidoptera, Noctuidae). [In
Russian with English summary.] Zoologicheskii Zhurnal 62: 354-360.

— 1984. Two new noctuid moth genera (Lepidoptera, Noctuidae). [In Russian.] Vestnik Zoologii 1984
(3): 73-74.

— 1985a. Phylogeny of the plusiine noctuid moths (Lepidoptera, Noctuidae). Part 1. Relationships of
the plusiine tribes. [In Russian.] Vestnik Zoologii 1985 (3): 10-16.

— 1985b. Phylogeny of the plusiine noctuid moths (Lepidoptera, Noctuidae). Part 2. Relationships of
the Palaearctic genera. [In Russian.] Vestnik Zoologii 1985 (4): 24-29.

Klots, A. B. 1970. Lepidoptera, pp. 115-130. In Tuxen, S. L. (ed.), Taxonomist’s glossary of genitalia in
insects. 359 pp. Copenhagen.

Kluge, A. G. & Farris, J. S. 1969. Quantitative phyletics and the evolution of the anurans. Systematic
Zoology 18: 1-32.

Kogi, H. 1986. Larva and food-plants of Diachrysia leonina (Oberthiir) and D. zosimi (Hiibner)
(Noctuidae: Plusiinae). Japan Heterocerists’ Journal 135: 145-146.

Kostrowicki, A. S. 1961. Studies of the Palaearctic species of the subfamily Plusiinae (Lepidoptera,
Phalaenidae). Acta zoologica cracoviensia 6: 367—472.

Kristensen, N. P. 1984. Studies on the morphology and systematics of primitive Lepidoptera (Insecta).
Steenstrupia 10: 141-191.

Laasonen, E. M. & Skvortsov, V. 1984. Notes on the immature stages and abundance of Autographa
excelsa (Lepidoptera, Noctuidae). Notulae Entomologicae 64: 164—166.

Lafontaine, J. D. In press. Journal of the Lepidopterists’ Society.

Lempke, B. J. 1965. Plusia tutti Kostrowicki and Plusia chrysitis L. (Lep., Noctuidae). Entomologische
Berichten. Amsterdam 25: 73-79.

— 1966a. Plusia chrysitis L. and Plusia tutti Kostrowicki Il (Lep., Noctuidae). Entomologische
Berichten. Amsterdam 26: 25-26.

— 1966b. Notes on the genus Autographa, subgenus Chrysaspidia Hiibner (Lep., Noctuidae). I. A.
“double” of Autographa (Chrysaspidia) festucae Linnaeus. Entomologische Berichten. Amsterdam 26:
64.

— 1969. Autographa festucae Linnaeus and Autographa gracilis Lempke. Entomologist’s Record and
Journal of Variation 81: 101-103.

Linnaeus, C. 1758. Systema Naturae Edn 10 Regnum Animale. 1: 824 pp. Holmiae.

McDunnough, J. H. 1916. On the types of certain noctuid genera occurring in North America (Lepidop-
tera). Entomological News 27: 393-400.

— 1938. Checklist of the Lepidoptera of Canada and the United States of America. Part 1. Macrolepi-
doptera. Memoirs. Southern California Academy of Sciences. 1: 1-272.

— 1944. Revision of the North American genera and species of the phalaenid subfamily Plusiinae
(Lepidoptera). Memoirs. Southern California Academy of Sciences. 2: 175-232.

Mell, R. 1943. Beitrage zur Fauna sinica. xxiv. Ueber Phlogophorinae, Odontodinae, Sarrothripinae,
“Westermannianae’ und Camptolominae (Noctuidae, Lepid.) von Kuangtung. Zoologische Jahrbiicher
76: 171-226.

Minet, J. 1982. Eléments sur la systematique des Notodontidae et nouvelles données concernant leur étude
faunistique a Madagascar. Bulletin de la Société Entomologique de France 87: 354-370.

Miyata, A. 1983. Handbook of the moth ecology. Moths as an indicator of the environment. [In Japanese
with English summary.] 1, 2. 48 + 1451 pp. Isahaya.

Mosher, E. 1916. A classification of Lepidoptera based on characters of the pupa. Bulletin of the Illinois
State Laboratory of Natural History 12: 17-159.

Moucha, J. & Novak, I. 1956. Kovoleskleci (Phytometrinae) Zapadnich Karpat. [Phytometrinae of the
Western Carpathians. ] (Lep. Phalaenidae). [In Czech with English summary.] Acta Faunistica Entomo-
logica Musei Nationalis Pragae 30: 53-63.

186 IAN J. KITCHING

Mukerji, S. & Singh, H. 1951. Studies on the chaetotaxy of larvae of Plusia species (Lepidoptera:
Phalaenidae). Proceedings of the Royal Entomological Society of London (B) 20: 15—24.

Mutuura, A., Yamamoto, Y. & Hattori, I. 1965. Early stages of Japanese moths in colour. 1. Revised by S.
Issiki. [In Japanese.] 127 pp. Osaka.

Nakamura, M. 1974. Pupae of Japanese Plusiinae (Lepidoptera: Noctuidae). Transactions of the Shikoku
Entomological Society 12: 1-9.

Neil, K. 1984. The larva of Autographa flagellum. (Walker) (Noctuidae: Plusiinae). Journal of the
Lepidopterists’ Society 38: 92-95.

Neilsen, E. S. 1985. A taxonomic review of the adelid genus Nematopogon Zeller (Lepidoptera:
Incurvarioidea). Entomologica Scandinavica (Supplement) 25: 1—66.

Nelson, G. J. 1974. Classification as an expression of phylogenetic relationships. Systematic Zoology 22:
344-359.

1979. Cladistic analysis and synthesis: principles and definitions, with a historical note on Adanson’s
Familles des plantes (1763-1764). Systematic Zoology 28: 1-21.

Nye, I. W. B. 1975. The generic names of moths of the World. 1: Noctuoidea (part): Noctuidae,
Agaristidae, and Nolidae. 568 pp. London.

Ochsenheimer, F. 1816. Die Schmetterlinge von Europa. 4. 223pp. Leipzig.

Oord, W. 1985. Geslaagde kweek va Macdunnoughia confusa (Stephens) (Lepidoptera: Noctuidae).
Entomologische Berichten. Amsterdam 45: 29-30.

Ottolengui, R. 1902. Plusia and allied genera with descriptions of new species. Journal of the New York
Entomological Society 10: 57-77.

Packard, A. S. 1869. The characters of the lepidopterous family Noctuidae. Proceedings of the Portland
Society of Natural History 1: 153-156.

Peigler, R. S. 1986. Worldwide predilection of resiniferous hostplants by three unrelated groups of moths
in the genera Actias, Citheronia (Saturniidae) and the subfamily Euteliinae (Noctuidae). Ty6 to Ga 37:
45-50.

Pinhey, E. C. G. 1975. Moths of Southern Africa. 273pp. Cape Town.

Prentice, R. M. 1962. Forest Lepidoptera of Canada. Canadian Department of Forestry Bulletin 128:
77-281.

Rath, O. vom. 1887. Uber die Hautsinnesorgane der Insecten. Zoologischer Anzeiger 10: 627-631,
645-649.

Rezbanyai, R. 1980. Diachrysia (Plusia) nadeja Oberthuer 1880 neu fuer die Schweiz und fuer Mitteleuro-
pa (Lepidoptera, Noctuidae). Entomologische Berichte Luzern 4: 26-27.

— 1983. Diachrysia chrysitis L. und D. nadeja Obth.- Beschreibung einer Parallel-Sommerzucht und der
Praimaginalstadien (Lep., Noctuidae). Mitteilungen der Schweizerischen Entomologischen Gesellschaft
56: 23-32.

1985. Diachrysia chrysitis (Linnaeus, 1758) und tutti (Kostrowicki, 1961) in der Schweiz. Ergebnisse
von Pheromonfallenfangen 1983—84 sowie Untersuchungen zur Morphologie, Phanologie, Verbreitung
und Oekologie der beiden Taxa (Lepidoptera, Noctuidae, Plusiinae). Mitteilungen der Schweizerischen
Entomologischen Gesellschaft 58: 345-372.

Richards, A. G. 1932. Comparative skeletal morphology of the noctuid tympanum. Entomologica
Americana 13: 1-43.

R.L. 1817. Naturgeschichte. [Book review of Ochsenheimer, F., 1816, Die Schmetterlinge von Europa 4.
223pp. Leipzig.] Jenaische Allgemeine Literatur-Zeitung 1: 273-293.

Robinson, G. S. 1975. Macrolepidoptera of Fiji and Rotuma, a taxonomic and geographic study. vii + 362
pp. Faringdon.

Roepke, W. 1941. On some new or little known Indo-Malayan noctuids (Lepidoptera Heterocera, family
Agrotidae). Zoologische Mededeelingen 23: 13-30.

Ronkay, L. 1982. Plusiinae (Lepidoptera, Noctuidae) from Korea. Folia Entomologica Hungarica 43:
137-145.

Rosen, D. E. 1978. Vicariant patterns and historical explanation in biogeography. Systematic Zoology 27:
155-188.

Ross, D. A. & Evans, D. 1957. Annotated list of forest insects of British Columbia. Part VI. Plusia spp.
(Noctuidae). Proceedings of the Entomological Society of British Columbia 54: 18.

Saunders, W. 1873. On the larva of Plusia balluca. Canadian Entomologist 5: 10-11.

Seitz, A. 1937. Subfamily: Stictopterinae. [General topics.] In Seitz, A., The Macrolepidoptera of the
World 11: 365—366. Stuttgart.

Sellier, R. 1975. Etude ultrastructural en microscopie electronique par balayage des organes sensoriels de
la trompe des Lépidoptéres Rhopalocéres. Alexanor 9: 9-15.

SYSTEMATICS OF THE PLUSIINAE 187

Sevastopulo, D. G. 1976. A list of the foodplants of East African Macrolepidoptera. Part 2 — Moths
(Heterocera) (part). Bulletin of the Amateur Entomologists Society 35: 177-195.

Skinner, B. 1984. Colour identification guide to moths of the British Isles (Macrolepidoptera). 267 pp.
Harmondsworth.

Skinner, H. 1917. New species. of Lepidoptera. Entomological News 28: 328-329.

_ Smith, J. B. 1882-1883. A synopsis of the North American genera of the Noctuidae. Bulletin of the

Brooklyn Entomological Society 4: 47-52; 5: 3-6, 11-14, 19-22, 27-30, 33-36, 43-46, 53-56.
— 1884. In Society news. Bulletin of the Brooklyn Entomological Society 7: 68.

— 1891. List of the Lepidoptera uf boreal America. v + pp. 6-124. Philadelphia.

_ South, R. 1961. The moths of the British Isles. 1. Edited and revised by H. M. Edelsten & D. S. Fletcher.

_ 427 pp. London.

Strand, E. 1917. Neue Aberrationen der Noctuiden — Subfamilien Hadeninae, Erastriinae, Catocalinae,

| Mominae und Phytometrinae. Archiv fiir Naturgeschichte 82 (A) 2: 28-50.

_ Sugi, S. 1972. The review of Dufay (1970) on Plusiinae from Madagascar. [In Japanese.]| Japan Heterocer-

| ists’ Journal 69: 145-146.

) —— 1982. Plusiinae. /n Inoue, H., Sugi, S., Kuroko, H., Moriuti, S. & Kawabe, A., The moths of Japan.
2. Plates and synonymic catalogue. pp. 386-388. Tokyo.

| Thorne, R. F. 1976. A phylogenetic classification of the Angiospermae. Evolutionary Biology 9: 35-106.

» —— 1983. Proposed new realignments in the angiosperms. Nordic Journal of Botany 3: 85-117.

| Tietz, H. M. 1972. An index to the described life histories, early stages and hosts of the Macrolepidoptera of

the continental United States and Canada. 2 vols. 1041 pp. Sarasota.

Tikhomirov, A. M. 1979. Phylogenetic relationships in the family Noctuidae (Lepidoptera) on the base of

| study of functional morphology of the male genital apparatus. Trudy Zoologicheskogo Instituta 82:
7-99.

| Timlin, J. S. 1955. The classification of the larval stage of some British Agrotidae. Unpublished Ph.D.

' thesis, University of London. 298 pp.

- Tutt, J. W. 1892. The British Noctuae and their varieties. 4, xxviii + 144 pp. London.

Tuxen, S. L. (ed.) 1970. Taxonomist’s glossary of genitalia in insects. 359 pp. Copenhagen.

|

Varley, G. C. 1962. A plea for a new look at Lepidoptera with special reference to the scent distributing
organs of male moths. Transactions of the Society for British Entomology 15: 29-40.

Warren, W. 1913. Subfamily: Phytometrinae. Jn Seitz, A., The Macrolepidoptera of the World. 3:
344-359. Stuttgart.

Warnecke, G. 1953. Zur Verbreitung von Phytometra bractea F. (Lep. Noct.) in Mitteleuropa. Entomolo-
gische Zeitschrift 63: 57-63.

Weidlich, M. 1982. Die Genitalunterschiede innerhalb der Gattung Abrostola O. sowie einige Bemer-
kungen zur Biologie von Abrostola asclepiadis Schiff. (Lep., Noct.). Entomologische Nachrichten 26:
165-168.

Westwood, J. O. 1838-1840. Synopsis of the genera of British insects. 158 pp. Jn: An introduction to the
modern classification of insects; founded on the natural habits and corresponding organisation of the
different families. 2 vols. London.

Wiley, E. O. 1979. An annotated Linnaean hierarchy, with comments on natural taxa and competing

| systems. Systematic Zoology 28: 308-337.
; a 1981. Phylogenetics. The theory and practice of phylogenetic systematics. 439 pp. New York.

Wiltshire, E. P. 1971. Osterreichische Expeditionen nach Persien und Afghanistan. Beitrage zur Lepidop-
terenfaunen, Teil 15 Noctuidae — Quadrifinae. Annalen des (K.K.) Naturhistorischen (Hof) Museums.
Wien 75: 627-649.

Wollaston, T. V. 1879. Notes on the Lepidoptera of St Helena, with descriptions of new species. Annals
and Magazine of Natural History; including Zoology, Botany and Geology (5) 3: 219-233, 239-343,
415-441.

_ Zimmerman, E. C. 1958. Insects of Hawaii. 7, Macrolepidoptera. ix + 542 pp. Honolulu.
:

) Appendix 1 Genera and type species of Plusiinae and outgroups.

The following is a list of the numbers of each sex examined in the Plusiinae and the outgroups. Shensiplusia
was scored from drawings published by Chou & Lu (1974), while Desertoplusia and Pseudochalcia were
described too recently to be included in the main analysis (but see Consensus classification).

Abrostola Ochsenheimer 1816
triplasia Linnaeus 1758 2 ©’, 2 (Fig. 32)

188 IAN J. KITCHING

Acanthoplusia Dufay 1970b

tarassota Hampson 1913 2 0’, 2 Q (Fig. 53)
Adeva McDunnough, 1944

albavitta Ottolengui 1902 1 o’, 1 9 (Fig. 71)
Agrapha Hibner [1821] 1816

ahenea Hubner [1821] 1816 2 co’, 2 9 (Fig. 57)
Allagrapha Franclemont 1964

aerea Hiibner [1803] 2 0’, 2 9 (Fig. 75)
Anadevidia Kostrowicki 1961

peponis Fabricius 1775 2 0’, 2 Q (Fig. 37)
Anagrapha McDunnough 1944

falcifera Kirby 1837 3 O’, 2 2 (Fig. 86)
Antoculeora Ichinosé 1973

ornatissima Walker 1858 2 ©’, 2 ° (Fig. 79)
Argyrogramma Hibner 1823

verruca Fabricius 1794 3 o’, 2 2 (Fig. 59)
Autographa Hibner [1821] 1816

bractea [Denis & Schiffermiiller] 1775 2 &’, 2 2 (Fig. 88)

gamma Linnaeus 1758 2 C’, 2 2 (Fig. 87)
Autoplusia McDunnough 1944

egena Guenée 18523 CO’, 2 2 (Fig. 80)
Caloplusia Smith 1884

hochenwarthi Hochenwarth 1785 2 ’, 2 2 (Fig. 95)
Chrysanympha Grote 1895

formosa Grote 1865 1 0’, 1 9 (Fig. 63)
Chrysodeixis Hibner [1821] 1816

chalcites Esper 1789 4 0’, 2 9 (Fig. 39)
Cornutiplusia Kostrowicki 1961

circumfiexa Linnaeus 1767 2 o’, 3 2 (Fig. 89)
Ctenoplusia Dufay 1970a

limbirena Guenée 1852 2 0’, 2 2 (Fig. 54)
Dactyloplusia Chou & Lu 1979

impulsa Walker 1865 2 CO’, 4 2 (Fig. 50)
Diachrysia Hubner [1821] 1816

chryson Esper 1789 2 0’, 2 2 (Fig. 84)
Eosphoropteryx Dyar 1902

thyatiroides Guenée 18522 0’, 2 Q (Fig. 61)
Erythroplusia Ichinosé 1962a

rutulifrons Walker 1858 4 0’, 2 9 (Fig. 73)
Euchalcia Hubner [1821] 1816

variabilis Piller 1783 2 0, 2 9 (Fig. 67)
Eutheiaplusia Dufay 1970a

pratti Kenrick 1917 1 & (holotype), 1 9 (Fig. 52)
Lamprotes R.L. 1817

c-aureum Knoch 1781 2 0’, 2 2 (Fig. 35)
Loboplusia Roepke 1941

vanderweelei Roepke 1941 2 & (holotype (Fig. 74) and paratype), 1 9 (holotype Autographa pokhara)

Lophoplusia Zimmerman 1958

pterylota Meyrick 1904 2 ©’, 2 9 (Fig. 92)
Macdunnoughia Kostrowicki 1961

confusa Stephens 18507 ©’, 3 9 (Fig. 76)
Mouralia Walker 1858

tinctoides Guenée 1852 2 0’, 2 P (Fig. 33)
Omorphina Alphéraky 1892

aurantiaca Alphéraky 1892 1 0’, 1 Q (Fig. 31)
Palaeographa Kljutshko 1984

interrogationis Linnaeus 1758 2 &’, 2 2 (Fig. 93)
Panchrysia Hubner [1821] 1816

deaurata Esper 1787 2 0’, 2 9 (Fig. 64)

SYSTEMATICS OF THE PLUSIINAE

Plusia Ochsenheimer 1816
festucae Linnaeus 1758 2 C’, 2 9 (Fig. 90)
Plusidia Butler 1879
cheiranthi Tauscher 1809 2 0’, 2 2 (Fig. 34)
Plusiopalpa Holland 1894
dichora Holland 1894 1 ©’, 2 9 (Fig. 58)
Plusiotricha Holland 1894
livida Holland 1894 1 o
Polychrysia Hiibner [1821] 1816
moneta Fabricius 1787 2 O’, 2 2 (Fig. 62)
Pseudeva Hampson 1913
purpurigera Walker 1858 2 CO’, 2 9 (Fig. 36)
Pseudoplusia McDunnough 1944
includens Walker 1857 4 CO’, 2 9 (Fig. 38)
Puriplusia Chou & Lu 1974
purissima Butler 1878 2 o’, 2 9 (Fig. 77)
- Rachiplusia Hampson 1913
nu Guenée 1852 4 Co’, 2 2 (Fig. 72)
_ Sclerogenia Ichinosé 1973
jessica Butler 1878 3 O, 2 P (Fig. 78)
Shensiplusia Chou & Lu 1974
__ nigribursa Chou & Lu 1974 [1 2]
Stigmoplusia Dufay 1970a
chalcoides Dufay 1968 1 0’, 1 P (Fig. 60)
Syngrapha Hibner [1821] 1816
___ devergens Hiibner [1813] 2 0’, 2 9 (Fig. 94)
Thysanoplusia Ichinosé 1973
intermixta Warren 1913 3 O’, 2 9 (Fig. 47)
_ Trichoplusia McDunnough 1944
___ ni Hubner [1803] 3 co’, 2 2 (Fig. 43)
Zonoplusia Chou & Lu 1979
ochreata Walker 1865 3 CO’, 2 9 (Fig. 49)

Plusiine taxa described too recently for inclusion in the analysis
Desertoplusia Kljutshko 1985
bella Christoph 1887 (Fig. 70)
| Pseudochalcia K\jutshko 1985
___inconspicua Graeser 1892 (Fig. 69)

Outgroups (nominal subfamily given in parentheses)
Anuga Guenée, 1852

constricta Guenée 1852 (Euteliinae) 2 0’, 2 9 (Fig. 101)
| Brachionycha Hiibner [1819] 1816
sphinx Hufnagel 1766 (Cuculliinae) 2 0’, 2 2 (Fig. 106)
Calophasia Stephens 1829 [June]
__lunula Hufnagel 1766 (Cuculliinae) 2 0, 2 2 (Fig. 104)
_Chalcopasta Hampson 1908
territans Edwards 1884 (Amphipyrinae) 2 0’, 2 2 (Fig. 102)
_Cucullia Schrank 1802
umbratica Linnaeus 1758 (Cuculliinae) 2 0, 2 2 (Fig. 105)
_Diloba Boisduval 1840

caeruleocephala Linnaeus 1758 (Dilobidae) 2 0’, 2 2 (Fig. 107)
Lophoptera Guenée 1852

squammigera Guenée 1852 (Stictopterinae) 2 0’, 3 (Fig. 99)
Magusa Walker 1857

orbifera Walker 1857 (Amphipyrinae) 2 0’, 2 9 (Fig. 97)
Oncocnemis Lederer 1853

confusa Freyer [1840] (Cuculliinae) 2 0’, 2 9 (Fig. 103)

189

190 IAN J. KITCHING

Paectes Hiibner 1818

cristatrix Guenée 1852 (Euteliinae) 3 0, 2 2 (Fig. 100)
Stictoptera Guenée 1852

cucullioides Guenée 1852 (Stictopterinae) 2 0, 4 9 (Fig. 98)

Appendix 2 Plusiine generic-level taxa not employed in this study.

Cerviplusia Chou & Lu 1974. Type species: Cerviplusia wukongensis Chou & Lu 1974. C. wukongensis was
synonymized with Antoculeora ornatissima (as Macdunnoughia ornatissima) by Dufay (1977). Thus,
Cerviplusia is a junior subjective synonym of Antoculeora Ichinosé.

Chrysaspidia Hiibner [1821] 1816. A junior objective synonym of Plusia Ochsenheimer.

Chrysoptera Berthold 1827. A junior objective synonym of Lamprotes R.L.

Cubena Walker 1856. Type species: Phalaena polydamia Stoll 1782. C. polydamia is a junior subjective
synonym .of Lamprotes c-aureum and has been treated as such for about a century. Cubena is thus a
junior subjective synonym of Lamprotes.

Inguridia Butler 1879. Type species: Inguridia abrostolina Butler 1879. A junior subjective synonym of
Abrostola Ochsenheimer. There is no doubt that abrostolina is congeneric with Abrostola triplasia and as
far as I am aware, /nguridia has not been used this century.

Melanoplusia Chou & Lu 1978. A junior objective synonym of Sclerogenia Ichinosé.

Neoplusia Okano 1963. Type species: Neoplusia furahatai Okano 1963. N. furahatai was synonymized with
Chrysodeixis acuta Walker by Dufay (1965), based on an examination of the genitalia. Neoplusia is thus a
junior subjective synonym of Chrysodeixis Hubner and has been treated as such since 1965.

Palaeoplusia Hampson 1913. Type species: Plusia venusta Walker 1865. McDunnough (1944) considered
P. venusta to be congeneric with Plusia festucae (as Chrysaspidia festucae). He therefore synonymized
Palaeoplusia with Chrysaspidia and thus the former became a junior subjective synonym of Plusia
Ochsenheimer. This decision has been upheld since and is followed in this study.

Paraplusia Mukerji & Krishnamorthy 1955. Type species: Plusia confusa Stephens 1850. This name is not
nomenclaturally available because it was published after 1930 without a statement of characters
differentiating the taxon.

Perloplusia Chou & Lu 1978. Type species: Perloplusia pseudopyropia Chou & Lu 1978. Chou & Lu
(1979a) subsequently considered this species to be synonymous with Erythroplusia pyropia Butler and
thus, by implication, Perloplusia is a junior subjective synonym of Erythroplusia (syn. n.).

Podioplusia Ichinosé 1962a. A junior objective synonym of Anadevidia Kostrowicki.

Scleroplusia Ichinosé 1962a. A junior objective synonym of Macdunnoughia Kostrowicki.

Unca Oken 1815. Included in a work rejected for nomenclatural purposes (ICZN, 1956).

Unca Lhomme [1929]. A junior objective synonym of Abrostola Ochsenheimer.

Yerongponga Lucas 1901. Type species: Yerongponga exequialis Lucas 1901. This genus and species was
described in the Plusidae [sic] at a time when Plusidae referred to the quadrifine noctuids in general
(Kitching, 1984). Yerongponga then, by inertia, became included in the Plusiinae (Nye, 1975). The
species was not examined in this study (the type specimen could not be located; it was searched for, but
not found, in the National Insect Collection, CSIRO, Canberra and in the South Australia Museum,
Adelaide). However, from the description, the species is evidently much larger than any known plusiine
and is most likely to be a species of Platyja Hiibner (J. D. Holloway, pers. comm.). Thus Yerongponga
can be transferred to the Catocalinae.

ee

i!

SYSTEMATICS OF THE PLUSIINAE 191

Appendix 3 Data matrix

Data set, characters 1-70, Plusiinae.

Omorphina 00000 01000 00101 00100 01010 01110 10000 00000 01000 0***0 01000 10010 00000 01000
Abrostola 11000 00001 00101 00101 02000 01010 10000 00000 00000 0**10 10000 10100 00000 01000
Mouralia 11000 00001 00122 00211 01000 01010 10000 00000 00000 0**10 10000 10100 00030 01000
Anadevidia 00010 00001 00111 00001 01000 01010 10000 00000 00000 0**10 00000 10110 00000 01000

Dactyloplusia 00110 10001 01121 00001 01000 01010 10000 00000 00000 0**10 00000 10111 00000 01000
Pseudoplusia 00110 10001 11121 00001 01000 01010 10000 00000 00000 0**10 00000 10100 00000 01000
Chrysodeixis 00110 10001 10121 00001 01000 01010 10000 00000 00000 0**10 00000 10110 00000 01000
Zonoplusia 00100 10001 01121 00001 01000 01010 10000 00000 00000 0**10 00000 10000 00000 01000
Trichoplusia 00110 10001 01121 00001 01000 01010 10000 00000 00000 0**10 00000 10101 00000 01000
Thysanoplusia 00010 10011 01121 00001 01000 01010 10000 00000 00000 0**10 00000 10101 00000 01000
Plusiotricha 00110 10001 101*1 00001 01000 01010 10000 00000 00000 0**10 00000 10102 10000 01000
Eutheiaplusia 00100 10001 00121 00001 01000 01010 10000 00000 00000 0**10 00000 10100 00000 01000

Agrapha 00110 00001 01121 00001 01000 01010 10000 00000 00000 0**10 00000 10100 00000 01000
Acanthoplusia 00120 10001 00121 00001 01000 01010 10000 00000 00000 0**10 00000 10101 00000 01000
Ctenoplusia 00130 10001 00121 00001 01000 01010 10000 00000 00000 0**10 00000 10102 00000 01000
Plusiopalpa 00110 10001 01121 00001 01000 01010 10000 00000 10000 0**10 00000 10100 10000 01000

Argyrogramma 00110 10001 00121 00001 01000 01010 10000 00000 00000 0**10 00000 10103 30000 01000
Stigmoplusia 00110 10001 10121 00001 01000 01010 10000 00000 00000 0**10 00000 10103 20000 01000
Shensiplusia 001** *0001 *11*1 *0**1 01000 01010 10000 000*0 00000 0**10 00000 10*** ***** 01000

Eosphoropteryx 00110 10001 00101 00101 01000 01010 10000 00010 00000 0**10 00000 10*10 00000 01000

Polychrysia 00111 10001 00101 00101 01000 01010 10000 00010 10000 0**11 00000 10110 00002 01000
Panchrysia 00010 10001 00101 10111 01000 01010 10000 00010 01000 0**10 00000 10110 00003 01000
Chrysanympha 00011 10001 00101 00101 01000 01010 10000 00010 11000 0**10 00000 10*10 00000 01000
Euchalcia 00040 00001 00101 00101 01000 01010 10000 00000 00000 0**10 00000 10010 00000 01000
Adeva 00041 00001 00101 00101 01000 01010 10000 00000 00000 0**10 00000 10*00 00000 01000
Plusidia 00010 00001 00101 00101 01000 01010 10000 00010 00000 0**10 00000 10010 00000 01000
Lamprotes 00110 10001 00101 10111 01000 01010 10000 00010 11000 0**10 00000 10110 00000 01000
Pseudeva 00011 10001 00101 10111 01000 01010 10000 00010 11000 0**11 00000 10110 00000 01000
Rachiplusia 00110 10001 00121 00001 01000 01010 10000 00000 00000 0**10 00000 10100 00120 01000
Erythroplusia 00111 00001 00101 00101 01000 01010 10000 00000 00000 0**10 00000 10110 00000 01000
Allagrapha 00010 10001 00101 10101 01000 01010 10000 00000 00000 0**10 00000 10010 00000 01000
Macdunnoughia 00110 10001 00101 00001 01000 01010 10000 00000 00000 0**10 00000 10110 00000 01000
Puriplusia 00110 10001 00101 00001 01000 01010 10000 00000 00000 0**10 00000 10110 00000 01000
Loboplusia 00110 00001 01101 00001 01000 01010 10000 00000 00000 0**10 00000 10100 00000 01000
Sclerogenia 00110 10001 01101 00001 01000 01010 10000 00000 00000 0**10 00000 10100 00000 01000
Autoplusia 00110 10001 00121 00001 01000 01010 10000 00000 00000 0**10 00000 10110 00010 01000
Antoculeora 00110 10001 01101 00101 01000 01010 10000 00000 00000 0**10 00000 10100 00000 01000
Diachrysia 00910 10011 00101 10101 01000 01010 10000 00000 00000 0**10 00000 10010 00001 01000
Anagrapha 00110 10001 00101 90101 01000 01010 10000 00000 00000 0**10 00000 10110 00020 01000
Autographa gamma 00110 10001 01101 00001 01000 01010 10000 00000 00000 0**10 00000 10110 00000 01000
A. bractea 00110 10001 00101 00101 01000 01010 10000 00000 00000 0**10 00000 10110 00000 01000
Cornutiplusia 00110 00001 01101 00101 01000 01010 10000 00000 00000 0**10 00000 10110 00010 01000
_ Plusia 00110 00001 00101 00101 01000 01010 10000 00000 00000 0**10 00000 10010 00000 01000
Lophoplusia 00110 00001 01101 00001 01000 01010 10000 00000 00000 21110 00000 10100 00003 01000
Palaeographa 00110 10001 01101 00101 01000 01010 10000 00000 00000 0**10 00000 10100 00010 01000
Syngrapha 00110 01001 01101 00101 01010 01010 10000 00000 00000 0**10 00000 10100 00010 01000

Caloplusia 00110 01001 01101 00101 01010 01010 10000 00000 00000 0**10 00000 10110 00220 01000

192

Data set, characters 71—150, Piusiinae.

Omorphin

Abrostol
Mouralia

Anadev
Dactylo
Pseudo
Chryso
Zono
Tricho
Thysano
P'tricha
Eutheia
Agrapha
Acantho
Cteno
P'palpa
Argyro
Stigmo
Shensi

Eosphor
Polychry
Panchry
Chrysan
Euchalc
Adeva
Plusid
Lamprot
Pseudev

Rachi
Erythro
Alla
Macdunn
Puri
Lobo
Sclero
Autoplus
Antocul

Diachry
Anagraph
A.gamma
A.bract
Cornuti
Plusia
Lopho
Palaeo
Syngraph
Caloplus

00000

00100
00100

00200
00200
00200
00200
00200
00200
00200
00200
00200
00200
00200
00200
00200
00200
00200
00200

00200
00200
00200
00200
00200
00200
00200
00200
00200

00200
00200
00200
00200
00200
00200
00200
00200
00200

00200
00200
00200
00200
00200
00200
00200
00200
00200
00200

00**0 KKKKK KKKKK KkKkKK

OO**1 **O**
00**1 **500

00**1 **300
11102 **200
10**1 KKK
13001 **500
10**1 **200
11002 **310
11002 **300
12002 **210
T2002) **0**
12012 **210
11002 **300
11202 **210
12002 **400
12332 **411
12022 **311

kKaeKKK KkKKK

00**1 KxKQ**
00**1 KkKOKK
00**]1 xk *K*
00**1 *kQ**
00**1 *xO*K*
00**] KKQ**
00**1 KkOK*
00**1 KkQK*
00**1 KKOK*

00**1 **300
00**1 **200
00**1 **200
00**1 **200
00**1 **200
O*x**] **200
00**1 **200
00**1 **300
OO**1"**200

00**1 **100
00**1 **100
00**1 **100
(lojra sseail(olo)
OOX*IS 1100
00**1 **300
00**1 **100
00**1 **100
00**1 xxx
00**1 **100

xkKQ*

00000

00010
00010
KKKKO
00010
00010
00010
00010
00010
***x10
11110
11010
01010
00010
10111
10011

KkKKK

akxQ*
KkKKO*
akxO*
*kKO*
KkKKO*
*kKQ*
KkKKQ*
KKKO*

KKKQ*

00000
00000
00000
00000
00000
0**00
00000
00000
00000

00000
00000
00000
00000
00000
00000
00000
00000

xkkO*

00000

00000
00000

00000
11120
00000
10110
10020
11220
11120
21220
KKKKO
21220
23220
23220
13220
22330
22330

Ka

00000
00000
00000
00000
00000
00000
00000
00000
00000

00000
00001
00001
00001
00001
00001
00000
00001
00001

00000
00000
00000
00000
00000
00000
00000
00000
00000
00000

IAN J. KITCHING

00000

00011
00011

00010
00010
00010
01010
00010
00010
00010
00010
00010
00110
00010
00010
00010
00110
01010

KeKKK

00010
00010
00010
00010
00010
00010
00010
00010
00010

00010
00010
00020
00020
00020
00010
00010
00010
00010

00010
00010
00010
00010
00010
00010
00010
01010
01010
01010

00000

01100
01100

00001
00001
00001
00000
00000
00001
00000
00000
00000
00001
00000
00000
00000
00001
00000

KkKKKK

10000
10000
10000
10000
10000
10000
10000
10000
10000

00000
00010
00000
00000
00000
*000*
00000
00000
00010

00000
00001
00001
00001
00001
00000
00001
00001
00001
00001

00000

01000
01000

10100
10000
10001
10001
10101
10100
10101
10001
10001
10101
10001
10101
10101
10001
10101

Kk

10000
10010
10010
11010
10010
10010
10010
10010
10010

10001
10100
10000
10000
10000
10000
10100
10000
10000

10000
10000
10110
10110
10110
10110
11110
10110
10110
10110

00000

00000
00000

00000
00000
10000
10000
00000
00000
01000
03000
00000
01000
01000
01000
01000
01000
12000

KKKKK

00000
00000
00000
00000
00000
00000
00000
00000
00000

00010
00000
00100
00100
00100
00100
00000
00000
00000

00000
00100
00000
00000
00000
00000
00000
00000
00000
00000

00100

001*0
00100

10100
00100
00101
00101
00100
00100
00100
00110
00100
00100
00110
00110
00100
00100
00100

KKK

01100
01100
01100
01100
01100
01100
01100
01100
01100

11100
11100
11100
11100
11100
*1100
11100
11100
00200

11100
11100
11100
11100
11100
11100
11100
11100
11100
11100

00000

2*011
00011

00000
00000
00000
00000
00000
00000
00000
00000
00000
00000
00000
00000
00000
20000
00000

KaKKK

10000
10000
10000
10000
10000
10000
10000
10000
10000

10000
10000
10000
10000
10000
100**
10000
10000
10000

10000
10000
10000
10000
10000
10000
10000
10000
10000
10000

01001

1000*
1000*

00001
00101
00200
01100
00001
0000*
00101
0*001
0*001
00000
00001
00001
00101
00000
00000

Kk KK

01001
01001
01001
01001
01001
01001
01001
01001
01001

01001
01001
00001
01001
01001
0**01
01001
0100*
01001

01001
02001
01001
01001
01001
01001
01001
02001
03001
03001

10000

0*010
0*010

10000
10000
10000
10000
10000
0*000
10000
10100
10000
10000
10100
10100
10000
10000
10000

KKK

10000
10000
11000
11000
10000
10000
10000
10000
10000

10000
10000
10000
10000
10000
10000
10000
0*000
10000

10000
10000
10000
10000
10000
10000
10000
10000
10000
10000

01000

00000
10000

10000
10000
10000
10000
10000
10000
10000
10101
10101
10100
10100
10100
10000
10000
10*00

KKK KK

00000
00000
00000
00000
00000
00000
00000
00000
00000

00010
00000
01010
01010
01010
000*0
00000
00000
00*0*

00000
00000
00000
00000
00000
00000
00000
00000
00000
00000

00310

001**
0020*

00310
00310
00300
00300
00310
00300
00300
00320
00310
00320
00300
00300
00300
00300
00310

kkk

00300
00300
00300
00300
00310
00310
00320
00310
00320

00300
10301
11300
11301
11301
00310
12300
02300
12301

00300
00210
00300
00300
00300
00300
00320
00310
00310
00310

*1100

00100
00100

000**
00110
00100
00100
00100
00110
00100
00100
00100
00110
00100
00100
00100
00100
00100

KKKKK

00100
*1100
*1100
*1100
*1100
*1100
*1100
*1100
*1100

oo0**
001**
00100
00102
00101
000**
000**
o00**
000**

1*100
1*100
1*100
1*100
1*100
1*100
1*100
1*110

1*110 |

1*110

SYSTEMATICS OF THE PLUSIINAE

Data set, characters 151—216, Plusiinae.

Omorphin

Abrostol
Mouralia

Anadev
Dactylo
Pseudo
Chryso
Zono
Tricho
Thysano
P'tricha
Eutheia
Agrapha
Acantho
Cteno
P'palpa
Argyro
Stigmo
Shensi

Eosphor
Polychry
Panchry
Chrysan
Euchalc
Adeva
Plusid
Lamprot
Pseudev

Rachi
Erythro
Alla
Macdunn
Puri
Lobo
Sclero
Autoplus
Antocul

Diachry
Anagraph
A. gamma
A.bract
Cornuti
Plusia
Lopho
Palaeo
Syngraph
Caloplus

00000

00000
00000

00000
00000
00000
00000
00000
00000
00000
00000
00000
00000
00000
00000
00000
00000
00000

KKK

00000
00000
00010
00010
00010
00010
00000
00000
00000

00010
00001
00000
00000
00000
0000*
00001
00000
00000

00000
00000
00000
00000
00000
00000
00000
00000
00000
00000

01000

01000
01000

02200
02200
02200
02200
02200
02200
02210
02211
02211
02200
02200
02200
02200
02200
02200

kkk RK

02100
02100
02100
02100
02100
02100
02100
02100
02100

02100
02200
02100
02200
02100
02100
02100
02000
02000

02100
02100
02100
02100
02100
02100
02100
02100
02100
02100

0Oo***

oo***
00O***

01000
oo***
oo***
oo**x*
01000
01000
01000
oo***
oo***
01000
oo***
oo***
oo***
01000
01000

kk

0oOo***
01001
01100
01100
01010
01010
01000
0oO***
01000

01000
oo***
oo***
oo***
oo***
oo***
oo***
oo***
oo***

01001
01000
01000
01000
01000
01000
01000
00***
oo***
0O***

*0000

*0021
*0011

00000
*1000
*1000
*1000
01000
01000
01000
*1000
*1000
01000
*1000
*1000
*1000
01000
01000

kak KKK

*0000
00000
00000
00000
00100
00100
00100
*0100
00100

00000
*0000
*0000
*0000
*0000
*0*0*
*0000
*0000
*0000

00000
00000
00000
00000
00000
10000
20000
*0000
*0000
*0000

00000

00000
00000

00**0
01110
01110
00110
00100
00100
00100
00100
00100
00100
00100
00100
00100
00100
00100

kk K RK

00001
00000
00**0
00**0
00000
00000
10**0
10**0
10**0

00000
00**0
00000
00**1
00000
00**1
00000
00000
00000

00000
00000
00000
00**0
00**0
00000
00000
00000
00000
00**0

00100

00100
00100

00100
00100
00100
00100
00100
00100
00100
Kk Kk
00100
00100
00100
00100
00100
00100
00100
**kKOO

00200
00200
00200
00200
00200
00200
00200
00200
00200

09100
00100
11100
11100
10100
01*00
11100
11100
11100

00100
00100
00100
00100
00100
00100
00100
00100
00100
00100

01010

01000
01000

01010
01011
01011
01011
01011
01011
01011
kk KK
01011
01011
01011
01011
01011
01011
01011
OLox*

01010
01010
01110
01110
01010
01110
01110
01110
01110

01010
01010
01010
01010
01010
010*0
01010
01010
01010

01010
01010
01010
01010
01010
01010
01010
01010
01010
01010

00000

01100
01100

00000
00000
00000
00000
00000
00000
00000
KkKKK
00000
00000
00000
00000
00000
00000
00000
*0000

00000
00000
00000
00000
00000
00000
10000
10000
10000

00000
00010
00010
00010
00010
00010
00010
00010
00000

00000
00010
00000
00000
00000
00000
00000
00010
00010
00010

00000

00000
10000

00000
00000
00000
00000
00000
00000
00000
KKK
00000
00000
00000
00000
00000
00000
00000
0*000

00000
00000
00000
00000
00000
00000
00020
10010
00020

10000
00100
10100
11100
12100
*0100
00100
00100
10100

10000
00000
00000
00000
00000
00000
00000
10000
10000
10000

00000

00000
00000

00000
00000
10000
10000
00000
00000
00000
KKK
00000
10000
01000
01000
00000
00000
00000
10*00

00000
10000
10000
10000
10000
10000
10000
10000
10000

00300
00000
00000
00000
00000
00000
00100
00211
00211

00000
00000
00000
00000
00000
00000
00000
00000
00000
00000

00000

00000
00000

00000
00000
00000
00000
00110
00211
00110
Kk
00111
00000
00000
00000
10000
00000
00000
001*0

00000
00000
00000
00000
10001
00001
00001
10000
00000

00000
00000
10000
00000
00000
00000
00000
00000
00000

00000
00000
00000
00000
00000
00000
00000
10000
10000
10000

00000

00000
00000

00010
00010
00000
01010
00010
00000
00000
KkKKK
00010
01000
00000
00000
00000
00000
00000
000*0

00000
00000
00010
00000
00000
00000
00000
00000
00000

10000
00010
00000
10000
10010
00010
00000
00000
00000

00000
00000
00000
00000
00000
00000
00000
00000
00000
00000

00200

00000

00000 0

00000
00000
00010
00000
00000
00000
00000
kK RK
00000
00010
00000
00010
00210
00010
00000
000*0

00100
00000
00000
00000
00000
00000
00000
00000
00000

00200
00000
00200
00300
00300
00200
01300
01300
01300

00000
00000
00000
00002
00003
00001
00001
00000
00000
00000

roe ero me 446.0. O Soe, Seo

fof MSV V SS) (= le loo Cr Sigh RS YON Co Tei) f=)

(Sy(el oy (so) (oy) (oute=y() (=)

194

IAN J. KITCHING

Data set, characters 1-70, 71-140, 141-210, 211-216, outgroups.

Magusa
Stictoptera
Lophoptera
Paectes
Anuga
Chalcopasta
Oncocnemis
Calophasia
Cucullia
Brachionycha
Diloba

Magusa
Stictoptera
Lophoptera
Paectes
Anuga
Chalcopasta
Oncocnemis
Calophasia
Cucullia
Brachionycha
Diloba

Magusa
Stictoptera
Lophoptera
Paectes
Anuga
Chalcopasta
Oncocnemis
Calopnasia
Cucullia
Brachionycha
Diloba

Magusa
Stictoptera
Lophoptera
Paectes
Anuga
Chalcopasta
Oncocnemis
Calophasia
Cucullia
Brachionycha
Diloba

00000
10000
00000
00000
10000
00000
00000
00000
00000
00000
00000

00001
01300
00300
11310
11310
00000
00000
00000
00002
10000
10000

010**
o00**
010**
000**
000**
001**
001**
010**
0020*
000**
100**

00000
10000
20000
10000
00000
00200
00000
00003
00000
00000
00200

00000
00100
00100
00000
00000
00011
00000
00000
00000
00000
00000

00**3
00**3
00**3
00**3
00**3
00**0
00**0
00**3
00**3
00**0
00**0

00100
000**
00100
000**
000**
00100
00100
00100
00100
00100
00100

SiS) (Sy feayrsy (Se) (Sy Sy (2S) (=)

00102
00050
00050
00040
00030
00002
00102
00103
00102
00122
00161

21***

O2***
0O1***
01***

1****
KKK

ka

30***
30***

Kak

KKK

00000
10100
10100
Lo) hale es
Ol
00000
00000
00000
00000
00000
00000

00312
01002
01002
00002
01002
00201
00201
00311
00111
00100
00000

Rk KKK
KaKKK
KaKKK
kkk
Ka KK
KKK
Ka
KKK
KKK
kak

Ka

10000
11000
11000
00000
10000
01000
01000
01000
01000
01000
01000

00100
00101
00101
00101
00101
00000
01000
01000
02000
10000
12000

KKK
KaKKKK
KKK
KKK
KKK
KaKKK
KKK
KARE
KKKKK
Kk KK

wKaeKKK

0O***
10***
10***
10***
10***
0O***
00***
00***
oo**x*
00***
00***

10100
10101
10100
10100
10101
00100
00100
01100
00100
00100
02100

00000
00000
00000
1***0
Q*kxO
01000
00000
1***0
00000
1***0
1***0

*0000
**000
**000
*0000
**000
*0000
*0000
*0000
*0000
*0000
*0000

21111
21111
21111
21212
21212
21101
21101
20101
20001
a2*2*
022%

00000
00000
00000
0**00
0**00
00001
00000
0**00
00000
0**00
0**00

00000
00000
00000
00000
00000
00000
00000
00000
00000
00000
00000

01000
01000
01000
11000
11000
00001
00001
00100
00101
00101
00101

0100*
0100*
01000
*0001
*0000
10000
00000
00000
00010
00000
00000

00000
00111
00111
00111
00111
00000
00000
00001
00000
10000
00000

00000
00101
00100
00101
00000
00010
01010
00000
00000
00010
00000

00000
000*0
00010
00011
00011
00000
00000
00000
00000
00000
00000

00000
10*00
10*00
00*00
00*00
00000
00000
00000
00000
00000
00000

0**00
0**00
0**00
10100
10100
0**10
0**00
0**00
0**00
22000
22000

00000
0***0
O***Q
00*00
0**00
00100
001*0
000*0
00100
00100
O***Q

00000
00000
00001
00001
00001
00000
00000
00000
00000
00000
00000

00000
00100
02000
02000
02100
00000
00000
00010
00010
00001
00001

00000
10100
10100
01100
01100
00000
3*000
3*000
00000
00000
**100

00000
00000
00001
00001
00001
00000
00000
00000
00000
10000
00020

10*00
11*00
11000
11100
11100
00110
00000
00000
00*10
00000
00*00

0000*
o000**
0000*
0001*
0001*
0000*
0000*
0000*
0010*
0000*
0000*

00000
01000
000*0
000*0
000*0
10000
00000
00000
10000
00000
10000

00000
12000
12000
01000
01000
00000
00000
00000
00000
00000
00000

0*001
0*000
0*000
0*000
0*000
0*001
0*001
0*001
0*001
0*000
0*000

00000
00000
01000
01000
01000
00000
00000
00000
00001
00000
00000

01000
11101
11001
11201
11201
01010
00010
01000
01000
01010
00000

00000
00000
00000
00000
00000
00000
00000
00000
00000
00000
00000

00001
00002
00022
00002
00002
00000
00010
00001
00011
00100
00100

SYSTEMATICS OF THE PLUSIINAE 195

Appendix 4 New taxa and synonymy, status changes and revivals.

New taxa
Autoplusiina subtrib. n.
New combinations (previous generic combination given in parentheses, genus of original description given
in square brackets).
Pseudoplusia dalei (Wollaston) (from ‘Plusia’ dalei) [ Plusia]
Agrapha meretricia (Schaus) (from ‘Plusia’ meretricia) {Plusia]
Agrapha calceolaris (Walker) (from ‘Plusia’ calceolaris) {| Plusia|
Autographa camptosema (Hampson) (from Macdunnoughia camptosema) [Phytometra]
Autographa schalisema (Hampson) (from Macdunnoughia schalisema) {[Phytometra}
New synonymy
Autographa emmetra Dufay, syn. n. of A. schalisema (Hampson)
Perloplusia Chou & Lu, syn. n. of Erythroplusia Ichinosé
Valid genera
Thysanoplusia gen. rey. (from synonymy with Trichoplusia)
Ctenoplusia gen. rev. (from synonymy with Agrapha)
Cornutiplusia gen. rev. (from synonymy with Syngrapha)
Lophoplusia stat. n. (from subgenus of Plusia s.1.)
Subgeneric changes
Pseudoplusia stat. n. — subgenus of Chrysodeixis
Eutheiaplusia stat. n. — subgenus of Plusiotricha
Acanthoplusia stat. rev. — subgenus of Ctenoplusia
Adeva stat. n. — subgenus of Euchalcia
Caloplusia stat. n. — subgenus of Syngrapha

Appendix 5 Generic classification of the Plusiinae

Subfamily Plusiinae
Tribe Omorphinini
Omorphina
Tribe Abrostolini
Abrostola
Mouralia
Tribe Argyrogrammatini
Argyrogrammatini incertae sedis: Shensiplusia
Anadevidia
Chrysodeixis
subgenus Chrysodeixis
subgenus Pseudoplusia stat. n.
Zonoplusia
Argyrogramma-complex
Trichoplusia, sedis mutabilis
Thysanoplusia genus-group, sedis mutabilis
Thysanoplusia stat. rev.
Dactyloplusia
Plusiotricha, sedis mutabilis
subgenus Plusiotricha
subgenus Eutheiaplusia stat. n.
Ctenoplusia, sedis mutabilis stat. rev.
subgenus Ctenoplusia stat. rev.
subgenus Acanthoplusia stat. rev.
Agrapha, sedis mutabilis
Plusiopalpa, sedis mutabilis
Argyrogramma genus-group, sedis mutabilis
Argyrogramma
Stigmoplusia

196 IAN J. KITCHING

Tribe Plusiini
Subtribe Euchalciina
Euchalciina incertae sedis: Pseudochalcia

Eosphoropteryx
Polychrysia
Panchrysia genus-group, sedis mutabilis
Panchrysia
Chrysanympha
Euchalcia genus-group, sedis mutabilis
Desertoplusia
Euchalcia
subgenus Euchalcia
subgenus Adeva
Lamprotes genus-group, sedis mutabilis
Plusidia
Lamprotes
Pseudeva
Plusiini incertae sedis: Diachrysia
Anagrapha
Rachiplusia

Subtribe Autoplusiina subtrib. n.
Autoplusiina incertae sedis: Erythroplusia
Loboplusia
Autoplusia genus-group
Sclerogenia
Autoplusia
Antoculeora
Macdunnoughia genus-group
Allagrapha
Macdunnoughia
subgenus Macdunnoughia
subgenus Puriplusia
Subtribe Plusiina
Autographa, sedis mutabilis
Plusia-complex, sedis mutabilis
Plusia genus-group
Plusia
Lophoplusia stat. n.
Cornutiplusia genus-group
Cornutiplusia stat. rev.
“‘Autographa”’ bractea
Syngrapha, sedis mutabilis
subgenus Palaeographa
subgenus Syngrapha
subgenus Caloplusia stat. n.

SYSTEMATICS OF THE PLUSIINAE 197

Figs 31-36 31, Omorphina aurantiaca (wingspan [ws] = 22 mm). 32, Abrostola triplasia (ws = 33 mm).
33, Mouralia tinctoides (ws = 45 mm). 34, Plusidia cheiranthi (ws = 30 mm). 35, Lamprotes c-aureum (ws
= 33 mm). 36, Pseudeva purpurigera (ws = 32 mm). Wingspans for Figs 31-107 are measured from
wingtip to wingtip and are given to the nearest mm.

198 IAN J. KITCHING

Figs 37-42 37, Anadevidia peponis (ws = 41 mm). 38, Chrysodeixis (Pseudoplusia) includens (ws = 37
mm). 39, Chrysodeixis (Chrysodeixis) chalcites (ws = 38 mm). 40, C. (C.) argentifera (ws = 35 mm). 41,
C. (C.) illuminata (ws = 39 mm). 42, C. (C.) kebea (ws = 34 mm).

SYSTEMATICS OF THE PLUSIINAE 199

Figs 43-48 43, Trichoplusia ni (ws = 35 mm). 44, T. sestertia (ws = 30 mm). 45, T. exquisita (ws = 35
mm). 46, T. lectula (ws = 30 mm). 47, Thysanoplusia intermixta (ws = 35 mm). 48, Thy. daubei (ws = 27
mm).

200 IAN J. KITCHING

Figs. 49-54 49, Zonoplusia ochreata (ws = 25 mm). 50, Dactyloplusia impulsa (ws = 24 mm). 51,
Plusiotricha (Plusiotricha) sp. near livida (ws = 31 mm). 52, Plusiotricha (Eutheiaplusia) pratti (ws = 36
mm). 53, Ctenoplusia (Acanthoplusia) tarassota (ws = 35 mm). 54, Ctenoplusia (Ctenoplusia) limbirena
(ws = 34 mm).

SYSTEMATICS OF THE PLUSIINAE 201

Figs 55-60 55, Ctenoplusia (Ctenoplusia) euchroa (ws = 41 mm). 56, ‘Plusia’ (s.1.) megaloba (ws = 28
mm). 57, Agrapha ahenea (ws = 37 mm). Note the well developed dorso-lateral hair tufts. 58,
Plusiopalpa dichora (ws = 43 mm). 59, Argyrogramma verruca (ws = 31 mm). 60, Stigmoplusia

| chalcoides (ws = 37 mm).

|

202 IAN J. KITCHING

Figs 61-66 61, Eosphoropteryx thyatiroides (ws = 38 mm). 62, Polychrysia moneta (ws = 39 mm). Note
the erect tegulae and scattered black scales on the forewing. 63, Chrysanympha formosa (ws = 33 mm).
64, Panchrysia deaurata (ws = 37 mm). 65, P. v-argenteum (ws = 35 mm). 66, P. dives (ws = 26 mm).

SYSTEMATICS OF THE PLUSIINAE 203

Figs 67-71 67, Euchalcia (Euchalcia) variabilis (ws = 38 mm). 68, ‘Plusia’ hampsoni (ws = 34 mm). 69,
_ Pseudochalcia inconspicua (ws = 35 mm). 70, Desertoplusia bella (ws = 28 mm). 71, Euchalcia (Adeva)
albavitta (ws = 27 mm).

204 IAN J. KITCHING

Figs 72-77 72, Rachiplusia nu (ws = 33 mm). 73, Erythroplusia rutulifrons (ws = 29 mm). 74, Loboplusia
vanderweelei (holotype) (ws = 36 mm). Note the enlarged anal fold on the hindwing, the large dorsal
hair tuft on the hindleg and the modified scales on the outer third of the hindwing. 75, Allagrapha aerea
(ws = 40 mm). 76, Macdunnoughia (Macdunnoughia) confusa (ws = 36 mm). 77, M. (Puriplusia)
purissima (ws = 32 mm).

SYSTEMATICS OF THE PLUSIINAE 205

Figs 78-83 78, Sclerogenia jessica (ws = 30 mm). 79, Antoculeora ornatissima (ws = 37 mm). 80,
Autoplusia egena (ws = 40 mm). 81, A. olivacea (ws = 43 mm). 82, ‘Syngrapha’ gammoides CO (ws = 46
mm). 83, ‘S.’ gammoides 9 (ws = 44 mm). Note the sexual dimorphism in ‘S.’ gammoides; the male is
grey and more contrastingly patterned than the brown female.

206 IAN J. KITCHING

Figs 84-89 84, Diachrysia chryson (ws = 40 mm). 85, D. chrysitis (ws = 36 mm). 86, Anagrapha falcifera
(ws = 35 mm). 87, Autographa gamma (ws = 40 mm). 88, ‘Autographa’ bractea (ws = 36 mm). 89,
Cornutiplusia circumflexa (ws = 38 mm).

SYSTEMATICS OF THE PLUSIINAE 207

Figs90-95 90, Plusia festucae (ws = 37 mm). 91, P. venusta (ws = 33 mm). 92, Lophoplusia pterylota (ws

= 38 mm). Note the strongly bipectinate antennae. 93, Syngrapha (Palaeographa) interrogationis (ws =
33 mm). 94, S. (Syngrapha) devergens (ws = 28 mm). 95, S. (Caloplusia) hochenwarthi (ws = 28 mm).

'

208 IAN J. KITCHING

Figs 96-101 96, Syngrapha (Syngrapha) parilis (ws = 26 mm). 97, Magusa orbifera (ws = 40 mm). 98,
Stictoptera cucullioides (ws = 35 mm). Note the bulbous eyes. 99, Lophoptera sp. near squammigera (ws
= 31 mm). 100, Paectes cristatrix (ws = 30 mm). Note the strongly bipectinate antennae. 101, Anuga
constricta (ws = 36 mm). Note the very weakly bipectinate antennae.

SYSTEMATICS OF THE PLUSIINAE 209

Figs 102-107 102, Chalcopasta territans (ws = 36 mm). 103, Oncocnemis confusa (ws = 28 mm). 104,
Calophasia lunula (ws = 29 mm). 105, Cucullia umbratica (ws = 50 mm). 106, Brachionycha sphinx (ws
= 41 mm). 107, Diloba caeruleocephala (ws = 37 mm).

210 IAN J. KITCHING

Figs 108-113 Probosces. 108, apex of proboscis, Autographa bractea. Note that the strengthening bars
remain semicircular to the apex and that there are few apical sensilla. 109, apex of proboscis, Anuga
constricta. Note that the strengthening bars break up apically into small plates and that there is a large
number of apical sensilla. 110, styloconic sensillum, Autographa bractea. 111, styloconic sensillum,
Cucullia umbratica. 112, styloconic sensilla, Oncocnemis confusa. Note the longitudinal ridges that are
absent in Autographa and Cucullia. 113, styloconic sensilla, Stictoptera cucullioides. Note the three
poorly developed ventral ridges. Scale as indicated.

SYSTEMATICS OF THE PLUSIINAE 211

Figs 114-119 Probosces. 114, styloconic sensillum, Paectes cristatrix. 115, styloconic sensillum, Anuga
constricta. 116, styloconic sensillum, Paectes cristatrix. Note on Figs 114—116, the three enlarged ‘wings’.
117, apex of proboscis, Cucullia umbratica. Note that the strengthening bars are broken up apically into
small nodules. 118, apex of proboscis, Anuga constricta. 119, base of proboscis, Anuga constricta. Note
the regular arrangement of the microtrichia. Scale as indicated.

DAD IAN J. KITCHING

120 121 122 123

| - |

124 125 Bs

2 Ee a 8
127 | 128

Figs 120-130 120-126, female frenula, right hindwings: (120) Anuga; (121) Paectes; (122) Stictoptera;
(123) Diloba; (124) Autographa gamma; (125) Anadevidia; (126) Thysanoplusia. 127, 128, right
hindwing apices: (127) Lamprotes; (128) Allagrapha. 129, 130, right forewing apices: (129) Pseudeva;
(130) Lophoplusia.

SYSTEMATICS OF THE PLUSIINAE 213

131

SSS

ee

137 138 139

Figs 131-139 131-134, right hindwing venation: (131) Paectes; (132) Autographa gamma; (133) Cucullia;
(134) Lophoptera. 135, 136, right forewing areoles: (135) Lophoplusia; (136) Lophoptera. 137-139,
right hindwing venation: (137) Syngrapha (Syngrapha); (138) Chalcopasta; (139) Calophasia, vein Mis
completely lost in this genus and is represented by a fold depicted as a dashed line.

214 IAN J. KITCHING

141

143

146

Figs 140-146 140-142, scale base patterns: (140) Omorphina; (141) Sclerogenia; (142) Paectes. 143,
tornal angle of Diloba, illustrating the apical remnant of vein CuP. 144-146, right pilifers, frontal view:
(144) Stictoptera; (145) ‘Autographa’ bractea; (146) Diloba. Scale line for Figs 140—142 = 0-1 mm and for :
Figs 144-146 = 0-2 mm.

SYSTEMATICS OF THE PLUSIINAE 215

151 =

Figs 147-152 Heads. 147, dorsal view, Mouralia. 148, lateral view, Mouralia. 149, dorsal view, Anuga.

150, lateral view, Anuga. 151, latero-frontal view, Syngrapha (Syngrapha). 152, frontal view, Stictop-
tera.

216 IAN J. KITCHING

aS ST
+
32 154
}
|
155
+
156
Sd

Figs 153-159 153-157, labial palps: (153) Oncocnemis; (154) Panchrysia; (155) Anagrapha; (156)
Chrysanympha; (157) Plusiopalpa. 158, 159, vom Rath’s organ tubes: (158) Stictoptera; (159) Abrostola.
Scale lines for Figs 153 and 155 = 0-5 mm and for Figs 158-159 = 0-1 mm.

SYSTEMATICS OF THE PLUSIINAE DAT)

Figs 160-167 160-163, antennal segments: (160) Lophoplusia; (161) Paectes; (162) Eosphoropteryx;
(163) Brachionycha. 164-167, left tegulae, dorsal view: (164) Abrostola; (165) Diloba; (166) Diachrysia;
(167) Brachionycha. The broad dorsal arm rests latero-dorsally on the mesothorax while the apically
microspinose ventral arm curves underneath the wings.

'
218 IAN J. KITCHING
'

171 172 173

176 177

SSS SSS SSS

Figs 168-177 168-170, ventral tegular arms, showing microspination: (168) Plusia; (169) Stictoptera;
(170) Cucullia. 171-173, outlines of ventral tegular arms: (171) Syngrapha (Palaeographa); (172)
Omorphina; (173) Paectes. 174, 175, outlines of dorsal tegular arms illustrating the marginal flange:
(174) Agrapha; (175) Cucullia. 176, 177, patagia: (176) Anuga; (177) ‘Autographa’ bractea. The | ,
attachment point (dashed outline) is ventral and medial, with the narrower angle (uppermost in these | >
drawings) directed laterally. Scale line for Figs 168-170 = 0-2 mm.

SYSTEMATICS OF THE PLUSIINAE 219

184

Figs 178-184 178-181, male hind-basitarsi illustrating the basitarsal combs: (178) Plusiopalpa; (179)
Thysanoplusia; (180) Ctenoplusia (Ctenoplusia); (181) Argyrogramma. 182-184, male fore-basitarsi:
(182) Plusiotricha (Plusiotricha) ; (183) Stigmoplusia; (184) Argyrogramma.

220 IAN J. KITCHING

Figs 185-190 185-187, hind-basitarsi: (185) Anagrapha; (186) Paectes; (187) Stictoptera. 188-190,
hind-tibiae: (188) outer view, Syngrapha (Syngrapha); (189) outer view, Rachiplusia; (190) inner view,
Mouralia.

SYSTEMATICS OF THE PLUSIINAE 221

193

194

Figs 191-204 191-194, fore-tarsi, third segment: (191) Lophoplusia; (192) Polychrysia; (193) Diachry-
sia; (194) Anagrapha. 195, 196, fore-tibiae: (195) Rachiplusia; (196) Syngrapha (Caloplusia). 197-200,
outlines of hind-tibiae, tibial spurs and basitarsi: (197) Anuga; (198) Stictoptera; (199) ‘Autographa’
bractea; (200) Euchalcia (Euchalcia). 201-204, tarsal claws: (201) Cucullia; (202) Autographa gamma;
(203) Plusiopalpa; (204) Oncocnemis. Scale line for Figs 201-204 = 0-5 mm.

222, IAN J. KITCHING
205
206
209 210
We

214
——

Figs 205-215 205-207, clawed fore-tibiae: (205) Brachionycha; (206) Chalcopasta; (207) Oncocnemis.
208, male St7, anterior edge uppermost, Anadevidia. 209, 210, lateral views from left of prothoracic
furcae: (209) Lophoplusia; (210) Anuga. 211-213, prothoracic furcae and epimera, posterior views:
(211) Lophoplusia; (212) Paectes; (213) Diloba. 214, 215, male St7s, anterior edges uppermost: (214)
Stigmoplusia; (215) Argyrogramma.

SYSTEMATICS OF THE PLUSIINAE 228

216

222
217

\ ) 7
218
219 ar cl
| aa zs

220

ae

aaa 224 225

Figs 216-229 216-221, internal flanges on the anterior edges of T2: (216) Cucullia; (217) Stictoptera;
(218) Lophoptera; (219) Omorphina; (220) Abrostola; (221) Anagrapha. 222, 223, St2’s, anterior edges
uppermost: (222) Syngrapha (Caloplusia); (223) Paectes psaliphora Hampson. 224, 225, female T7’s,
posterior edges uppermost: (224) Adeva; (225) Autoplusia egena. 226-229, diagrammatic cross-sections
of dorso-lateral scent patches on male T7’s; scale arbitrary: (226) Agrapha; (227) Dactyloplusia; (228)
Ctenoplusia (Ctenoplusia) ; (229) Argyrogramma.

224 IAN J. KITCHING

Figs 230-234 Counter tympanal hoods and pleural pouches. The apodemes from St2 and the post-
spiracular bars are shaded. 230, anterior view, RHS, Thysanoplusia orichalcea. 231, lateral view, RHS,
Thysanoplusia orichalcea. 232, anterior view, LHS, Panchrysia. 233, anterior view, RHS, Stictoptera.
234, anterior view, RHS, Calophasia.

SYSTEMATICS OF THE PLUSIINAE 225

239 240
4

Figs 235-240 St8 and sternal hair pencils, ventral views. 235, Brachionycha, hair pencils absent. 236,
Cucullia, hair pencil represented by a sparse tuft of scales arising from a sclerotized line. 237, Magusa,
the lateral arms of St8 have been flattened so that the hair pencils have been turned outwards through
90°. 238, Paectes. 239, Stictoptera. 240, Anuga, the hair pencil pockets have been turned through 180° so
as to face anteriorly instead of posteriorly. In Figs 237 and 238, and in subsequent illustrations of hair
pencils, the pencils themselves are not drawn in detail but their limits are shown by dashed lines. If limits
are not shown, this indicates that the hair pencil scales had become detached before drawing. In
addition, the shading at the base of the pockets indicates the extent of the main fields of hair pencil scale
insertions.

226 IAN J. KITCHING

244

Figs 241-247 St8 and sternal hair pencils, ventral views. 241, Lamprotes. 242, Chrysodeixis (Chry-
sodeixis) argentifera. Note the expanded secondary hair brushes, the spatulate apical scales and the
complete fusion of the hair pencils and their pockets. 243, Macdunnoughia (Macdunnoughia). 244,
Autoplusia olivacea. 245, ‘Autographa’ bractea. 246, Plusiotricha (Eutheiaplusia). Note the reduced hair
pencils and reduced secondary hair brushes. 247, Chrysodeixis (Pseudoplusia). Note the total loss of all

hair pencils and brushes.

SYSTEMATICS OF THE PLUSIINAE 227

250

252

Figs 248-253 248, 249, St8 and hair pencil of Mouralia: (248) ventral view; (249) lateral view. 250-253,
A8 and hair pencil structures of Chrysodeixis (Chrysodeixis) chalcites: (250) St8, posterior edge
uppermost; (251) lateral view of A8 and hair pencil complex; note how the sagittate St8 supports the
almost spherical hair brush ventrally, while T8 supports it dorsally, with the lateral arms of T8 curving
laterally and ventrally to support the mouth of the hair brush pocket; (252) T8, posterior edge
uppermost; (253) anterior view of A8 and the hair pencil complex, showing how the anterior part of St8
supports the bases of the hair brushes.

228

IAN J. KITCHING

Figs 254-260 St8 and sternal hair pencils. 254, Trichoplusia, ventral view. 255, Trichoplusia, lateral view.
256, Thysanoplusia, ventral view. 257, Thysanoplusia, lateral view. 258, Argyrogramma, ventral view;
note the differentiation of the hair pencil scale insertions and the large semicircular flange of scales
attached to the apex of the dorsal arms of St8. 259, Argyrogramma, anterior view of left hair pencil
pocket, showing the different fields of hair pencil scale insertions, dorsal side is uppermost. 260,

Argyrogramma, lateral view; note the long, thin, curved scales arising from the spatulate area of scale
bases and projecting beyond the main hair pencils.

SYSTEMATICS OF THE PLUSIINAE 229

Figs 261-266 St8 and sternal hair pencils. 261, Plusiotricha (Plusiotricha), ventral view. 262, Plusiotricha
(Plusiotricha), lateral view; note the large secondary hair brushes. 263, Ctenoplusia (Acanthoplusia),

ventral view. 264, Ctenoplusia (Acanthoplusia), lateral view. 265, Plusiopalpa, ventral view. 266,
Plusiopalpa, lateral view.

230 IAN J. KITCHING

Figs 267-271 St8 and sternal hair pencils. 267, Ctenoplusia (Ctenoplusia), ventral view. 268, Ctenoplusia
(Ctenoplusia), lateral view; note that the outer hair pencil scales are shorter than the inner scales. 269,
Stigmoplusia, ventral view; note the large semicircular flange of scales attached to the dorsal arms of St8.
270, Stigmoplusia, anterior view of left hair pencil pocket; note the differentiated fields of hair pencil
scale bases. 271, Stigmoplusia, lateral view.

SYSTEMATICS OF THE PLUSIINAE 2B

a

@ os

Figs 272-278 272, 273, St8 and sternal hair pencils, Agrapha: (272) ventral view; (273) lateral view; note
the differentiated hair pencils, similar to the condition seen in Argyrogramma (Figs 258—260). 274-279,
unci: (274) Abrostola, lateral view; (275) Abrostola, dorsal view; (276) Anuga, lateral view; (277)
Anuga, lateral view; (278) Mouralia, dorsal view; (279) Mouralia, lateral view. In Figs 274 and 279, the
LAST’s are shaded, while the shading on Fig. 278 indicates the field of long, deciduous hairs that arise on
the uncus of this genus.

237) IAN J. KITCHING

281

280 282

Figs 280-286 Unciandtegumena. 280, Stigmoplusia, dorsal view. Note the lateral flanges that are unique
to certain members of this genus. 281, Stigmoplusia, lateral view. 282, Oncocnemis, lateral view. 283,
Syngrapha (Caloplusia) , lateral view. 284, Trichoplusia, lateral view. 285, Euchalcia (Euchalcia), lateral
view. 286, Anuga, dorso-lateral view. Scale line for Fig. 286 = 0-5 mm.

SYSTEMATICS OF THE PLUSIINAE 233

f-O_N

288

Figs 287-291 287-290, unci and tegumena: (287) Macdunnoughia (Macdunnoughia), lateral view; (288)
Anadevidia, dorsal view; (289) Chrysodeixis (Chrysodeixis), lateral view; the LAST has been shaded;
(290) Anagrapha, dorsal view. 291, lateral view of the terminal segments of Paectes, to illustrate the
setose socii and the beak-like gnathos. The valves have been removed.

234 IAN J. KITCHING

294

295 ee A

}-—_______
297

Figs 292-298 292, lateral view of the terminal segments of Anuga, to show the reduced uncus and large,
curved, apically hooked socii. 293, 294, lateral views of unci and tegumena: (293) Argyrogramma; (294)
Euchalcia (Adeva). 295-297, dorsal views of tegumena: (295) Erythroplusia; (296) Allagrapha; (297)
Antoculeora.

SYSTEMATICS OF THE PLUSIINAE 235

pe

299

Figs 298-305 Vincula. 298, Agrapha, ventral view. 299, Stigmoplusia, dorso-lateral view. 300, Stigmo-
plusia, ventral view. 301, Thysanoplusia, ventral view. 302, Rachiplusia, ventral view. 303, Autographa
gamma, ventral view. 304, Plusiotricha (Plusiotricha), ventral view. 305, Plusiotricha (Plusiotricha) ,
dorso-lateral view.

236 IAN J. KITCHING

Figs 306-312 306, 307, pleurite of Ctenoplusia (Ctenoplusia), to illustrate its attachment to the vinculum
and tegumen: (306) ventro-lateral view; (307) lateral view. 308-310, ventral view of vincula: (308)
Chrysodeixis (Chrysodeixis); (309) Lophoptera; (310) Paectes. 311, 312, lateral views of vincula and
pleurites: (311) Rachiplusia; (312) Anadevidia.

SYSTEMATICS OF THE PLUSIINAE 235),

315

316

320 321

Figs 313-321 Lateral views of pleurites and their attachments to their vincula and tegumena. 313,
Mouralia. 314, Magusa. 315, Calophasia. 316, Antoculeora. 317, Oncocnemis. 318, Abrostola. 319,
‘Autographa’ bractea. 320, Diloba. 321, Argyrogramma.

238 IAN J. KITCHING

328

331

332

ee 334

Figs 322-334 322, 323, diagrammatic lateral views showing the relationship between the apex of the juxta
and that of the aedeagus: (322) Abrostola; (323) generalized member of the Autoplusiina. 324-334,
juxtae: (324) Mouralia, ventral view; (325) Abrostola, ventral view; (326) Sclerogenia, ventral view;
(327) Sclerogenia, lateral view; (328) Dactyloplusia, ventral view; (329) Thysanoplusia, ventral view;
(330) Plusiopalpa, ventral view; (331) Panchrysia, ventral view; (332) Chrysodeixis (Chrysodeixis)
argentifera, lateral view; (333) Paectes, ventral view; (334) Chrysodeixis (Chrysodeixis) chalcites, ventral
view. On Figs 326-331 and, subsequently, Figs 335-340, black shading at the base of the juxta indicates
the area where it is fused to the valve flanges.

SYSTEMATICS OF THE PLUSIINAE 239

337

338 339

335

| | E>
340 341

343

346
345

Figs 335-346 335-341, juxtae: (335) Syngrapha (Palaeographa), ventral view; (336) Syngrapha
(Palaeographa), lateral view; (337) Syngrapha (Syngrapha), ventral view; (338) Syngrapha (Caloplu-
sia), ventral view; (339) Syngrapha (Caloplusia), lateral view; (340) Anagrapha, ventral view; (341)
Chrysodeixis (Pseudoplusia), ventral view. 342, ventral view of the juxta and the bases of the valves of
Agrapha, to illustrate the U-shaped juxta and the basal valve flanges. 343, base of the valves and juxta of
Panchrysia, to show the bulge that appears between the valve bases when the valves are opened flat.
344-346, postero-lateral views of transtillae: (344) Plusiopalpa; (345) Erythroplusia; (346) Euchalcia
(Euchalcia).

344

240 IAN J. KITCHING

: ”

350

Figs 347-350 Left valves. 347, Abrostola. 348, Mouralia; note the apical twist to the valve, bearing a large
fan of spatulate scales, the limits of which are indicated by the dashed line. 349, Cucullia. 350,
Polychrysia.

SYSTEMATICS OF THE PLUSIINAE 241

Figs 351-359 351-355, left valves: (351) Plusiotricha (Plusiotricha); (352) Plusiotricha (Eutheiaplusia) ;
(353) Stigmoplusia; (354) Agrapha; the dashed line shows the limits of the short, curved, upright scales
on the flattened edge of the sacculus; (355) Ctenoplusia (Acanthoplusia); the dotted oval indicates the
position of the pad of inflated scales on the external surface of the valve. The small inverted ‘u’s on the
apex of the valves in Figs 351-355 (and also on Figs 367 and 379) are the insertions of the basally directed
apical scales found in the members of the Argyrogrammatini. The dark shading at the bases of several
valves indicates to points of fusion with the juxta (cf. Figs 326-331 and 335-340). 356-359, clavi: (356)
Oncocnemis; (357) Cucullia; (358) Antoculeora; (359) Sclerogenia. Scale line for Fig. 354 = 0-5 mm.

242 IAN J. KITCHING

366 pees feemers teete 367

Figs 360-367 360-361, left valves: (360) Rachiplusia; (361) Sclerogenia. 362-365, subapical cornuti:
(362) Euchalcia (Euchalcia), dorsal view; (363) Euchalcia (Euchalcia), lateral view; (364) Euchalcia
(Adeva), dorsal view; (365) Euchalcia (Adeva), lateral view. 366, left valve and juxta, Diachrysia. 367,
left valve, Dactyloplusia. Scale lines for Figs 362—365 = 0-5 mm.

SYSTEMATICS OF THE PLUSIINAE 243

| Figs 368-371 Left valves. 368, Allagrapha. 369, Macdunnoughia (Puriplusia). 370, Macdunnoughia
(Macdunnoughia); the black points are heavily sclerotized. 371, Loboplusia.
| ;

244 IAN J. KITCHING

Figs 372-374 Left valves. 372, Autoplusia egena. 373, Autoplusia olivacea. 374, ‘Syngrapha’ gammoides.
Note in the last species that the juxta is fused with the clavus, which has lost its connection with the valve
compietely.

SYSTEMATICS OF THE PLUSIINAE 245

Figs 375-378 375-377, left valves: (375) Paectes; (376) inner view of Antoculeora; (377) dorsal view of
Antoculeora. 378, both valves and juxta of Erythroplusia; note the asymmetrical harpes.

246 IAN J. KITCHING

Figs 379-385 379-382, left valves: (379) Trichoplusia; (380) Syngrapha (Caloplusia); (381) Autographa
gamma; (382) Lophoplusia. 383-385, auricular cornuti; (383) Loboplusia; (384) Macdunnoughia
(Macdunnoughia); (385) Eosphoropteryx. Scale lines for Figs 380 and 383-385 = 0-5 mm.

SYSTEMATICS OF THE PLUSIINAE 247

wat

396
393 394 395 || a SS

| Figs 386-396 386, 387, left valves: (386) Syngrapha (Palaeographa); (387) Stictoptera. 388, 389, vincula
| and valves, outer view: (388) Paectes; (389) Euchalcia (Euchalcia). 390-392, diagrammatic lateral views
| of aedeagi illustrating the three states of character 157: (390) state 0; (391) state 1; (392) state 2. 393-395,

diagrammatic ventral views of aedeagi to illustrate the three states of character 158: (393) state 0; (394)
| state 1; (395) state 2. Scale for Figs 390-395 is arbitrary. 396, dorsal view of the basal part of the vesica of
Pseudeva, to illustrate the patch of curved cornuti.

248 IAN J. KITCHING

Figs 397-402 Aedeagi and vesicas. 397, Chrysodeixis (Pseudoplusia), lateral view. 398, Mouralia,
ventral view. 399, Mouralia, lateral view. 400, Plusiopalpa adrasta, ventro-lateral view. 401, Dactyloplu-
sia, lateral view. 402, Syngrapha (Palaeographa). The dashed ovals on Figs 400 and 402 represent the
opening of the cuticular simplex. The dashed lines on Figs 397 and 401 represent the internally
invaginated sacs.

SYSTEMATICS OF THE PLUSIINAE 249

| 407

. 406
408
———eEe————
HW]
409
410
(7s Saas
411
K——______

413

L or

Figs 403-413 403-405, subapical cornuti: (403) ‘Autographa’ bractea; (404) Plusia; (405) Lophoplusia.
406, diagrammatic representation of the serpentine-coiled vesica of the genus Plusiotricha; scale
arbitrary. 407, Abrostola, aedeagus, ventral view. 408, 409, subapical cornuti: (408) Chrysanympha;
(409) Panchrysia. 410, 411, aedeagi: (410) Ctenoplusia (Acanthoplusia), lateral view; (411) Plusidia,
lateral view. 412, 413, anal papillae and intersegmental membrane: (412) Anadevidia, showing the
membranous pouch anterior to the anal papillae; (413) Cucullia, showing the unexpanded membrane
anterior to the anal papillae.

250 IAN J. KITCHING

Figs 414-420 Anal papillae. 414, Ctenoplusia (Acanthoplusia). 415, Thysanoplusia. 416, Anadevidia,
417, Paectes, postero-lateral view. 418, Lophoptera, postero-lateral view. 419, Panchrysia. 420,
Cucullia.

SYSTEMATICS OF THE PLUSIINAE 251

—— 7

‘a
us

Figs 421-430 421-426, female segments A8: (421) Abrostola, ventral view; (422) Abrostola, lateral view;
(423) Mouralia, lateral view; (424) Plusiopalpa, lateral view; (425) Anagrapha, lateral view; (426)
Paectes, lateral view. 427, 428, antra: (427) Syngrapha (Palaeographa), ventral view; (428) Rachiplusia,
ventral view. 429, 430, female segments A8, lateral views: (429) Anuga; (430) Euchalcia (Adeva).

DP IAN J. KITCHING

Mis

431

Figs 431-437 Antra. 431, Macdunnoughia (Macdunnoughia), ventral view. 432, Macdunnoughia (Puri-
plusia), ventral view. 433, Sclerogenia, ventral view, illustrating the tuft of erect scales at the mouth of
the antrum. 434, Plusidia, ventral view. 435, Lamprotes, ventral view. 436, Antoculeora, lateral view,
right. 437, Antoculeora, ventral view; note the large, asymmetrical, sclerotized, open antrum.

SYSTEMATICS OF THE PLUSIINAE 253

443

444

Figs 438-444 438-440, A8 and antrum of Autoplusia olivacea: (438) postero-ventral view; (439) ventral
view; (440) lateral view, left. 441, dorsal view of the flattened section with recurved edges of the ductus
bursae of Macdunnoughia (Macdunnoughia). 442-444, ducti and corpora bursae: (442) Dactyloplusia,
ventral view; (443) Argyrogramma, lateral view; (444) Agrapha.

254 IAN J. KITCHING

446

447

448

Figs 445-448 Ducti and corpora bursae. 445, Rachiplusia. 446, Thysanoplusia. 447, Trichoplusia. 448,
Plusiotricha (Eutheiaplusia).

SYSTEMATICS OF THE PLUSIINAE 255

453

Figs 449-453 Ducti and corpora bursae. 449, Allagrapha. 450, Macdunnoughia (Macdunnoughia). 451
Macdunnoughia (Puriplusia). 452, Eosphoropteryx. 453, Erythroplusia.

>

256 IAN J. KITCHING

- HN

456

Figs 454-456 Ducti and corpora bursae. 454, Autoplusia. 455, Sclerogenia. 456, Antoculeora. The dashed
line indicates the discontinuity in ornamentation of the corpus bursae (character 212).

SYSTEMATICS OF THE PLUSIINAE 257)

Figs 457-460 Ducti and corpora bursae. 457, distal part of the corpus bursae of Paectes, illustrating the
concave signum and the circular ridges. 458, Cornutiplusia, the dashed lines show the extent to which the
ductus sclerotization continues onto the apex of the corpus bursae. 459, Anagrapha. 460, Lophoptera.

258 IAN J. KITCHING

461 462

464

Figs 461-465 461-463, ducti and corpora bursae: (461) Plusia; (462) Lophoplusia; (463) ‘Autographa’
bractea. 464, 465, corpora bursae: (464) Calophasia, showing the sclerotized longitudinal ridge; (465)
Lophoptera, illustrating the invaginated, spinose signum.

Abrostola 101 102 103 140 177 178
180 190

abrostolina 141 190

Abrostolini 84 87 140 175 179

abrostoloides 124

abstrusa 166

acalypta 128 134 149

Acanthoplusia 113 133 147 171

accentifera 139 147

acuta 92 132 143 145 172 190

Adeva 97 107 115 151 152 174

adiaphora 147

adrasta 94 148

aemula 81 162

aenofusa 128 129 134 135 149

aerea 7991 155

aereoides 85 155 156 160

agnata 81 147 149 171

Agrapha 79 113 146

ahenea 79 146

ain 166 167

albavitta 152

albostriata 81 128 134 144 148 177

alias 166

Allagrapha 79 106 117 155

allocota 128 134 149

alpina 165

altera 164

alticola 133 166 180

ampla 79 162

amydra 147

Anadevidia 119 142 153 160 177
178 190

Anagrapha 103 106 115 159 161
164

angulidens 133 166

angulum 92

Anigraea 169

anophioides 141 178

Antoculeora 93 107 117 119 158
177

antsalova 149

Anuga 94 124 169

arachnoides 92 128

argentifera 93 95 132 143 179

Argyrogramma 107 113 148

Argyrogrammatini 84 87 142 175
179

argyrosigna 81

asclepiadis 141

augusta 136 150

aurantiaca 140

aurata 150

aureus 150

Autographa 115 156 157 161 162
177 180

Autoplusia 93 117 154 158

Autoplusiina 153 175

balluca 85 155 160
basigera 91 128 149
bella 172

bellieri 152

SYSTEMATICS OF THE PLUSIINAE

Index

bieti 142 156 160

biloba 79 91 161

bimaculata 161

bipartita 145

borea 125 164

Brachionycha 94 108 119 169 180
bractea 81 115 162 177

brassicae 145

brevistriata 144

buraetica 180

c-aureum 153

calceolaris 146

californica 79 161 180

callista 128

Calophasia 94 105 119

Caloplusia 107 115 154 164 166

Caloplusiini 87

camptogamma 148

camptosema 156

carcassont 97 146

cashmirensis 151 152

celsa 79 165 166 177

Cerviplusia 190

chalcites 93 96 132 143 145

chalcoides 149

Chalcopasta 94 103 105 108 119

cheiranthi 152

chinghaiensis 152

chlorocharis 152

Chrysanympha 97 103 124 151 177

Chrysaspidia 190

chrysitis 78 83 124 139 155 160

Chrysodeixis 93 95 102 111 143 172
179 190

chryson 83 106 160

Chrysoptera 190

circumflexa 92 162

circumscripta 81 149

composita 166 167

confusa 156 157 190

consona 152

contexta 163

coreae 160

Cornutiplusia 106 115 162

corusca 79 162

crassisigna 135 156 157

cristatrix 94

cryptica 166

Ctenoplusia 113 133 145 147 148
177

Cubena 190

Cucullia 94 100 101 106 107 112
119

cuprea 152

Dactyloplusia 102 143 145

dalei 143

daubei 81 124 144 145

deaurata 78 151

Desertoplusia 172

devergens 125 133 165 166 167 180
Diachrysia 83 102 155 159 178

259

diasema 125 161 164
dichora 93 103 148 180
Diloba 94 103 109 119 169
diversigna 125 164

dives 125 151

dolera 128

dorsiflava 136 151
dyscapna 146

egena 93 129 134 139 159

egenella 159

egenoides 158

ekekei 145

emmetra 156

Eosphoropteryx 150

epargyra 134

epigaea 164

epistilba 149

eriosoma 143

Erythroplusia 117 154 177 178 190

esmeralda 150 151

Euchalcia 106 115 151 152 174 177
180

Euchalciina 149 153 175

Euchalciini 87

eugrapha 147

Eutelia 169

Euteliinae 119 167 168 169

Eutheiaplusia 97 101 102 108 146

excelsa 162

exequialis 190

exquisita 125 128

falcifera 85 161
festucae 78 83 139 163
flagellum 161
flammifera 165
fletcheri 97 146
florina 128 145
formosa 78 151
furahatai 190
furcifera 147

gamma 115 139 145 161
gammoides 93 134 154 159
gerda 78 90 152

giffardi 164

gilarovi 165

gorilla 144

gromieri 128

hainanensis 149

hampsoni 151

hebetata 134 142

herbuloti 147

hildebrandti 148
hochenwarthi 125 133 166 167
homoia 145

ichinosei 147

ignea 125 133 165 166 167
ignicollis 145

illuminata 93 132

260

illustrata 159

imperatrix 124 152
impulsa 143

includens 91 139 143 145
inconspicua 151 174
indicator 124

Inguridia 190

intermixta 81 144 145
interrogationis 165 180
italica 136 152

Jessica 158
Jota 162

kaszabi 140
kebea 93 96 132
kostjuki 162

Lamprotes 106 107 115 153
latistigma 147

lavendula 78

lenzeni 151

leonina 124 142 155 156 160
leucostigma 128 134
limbirena 92 114 124 128 147
livida 97 146

Loboplusia 94 97 117 157
locuples 93 158

longisigna 144

Lophoplusia 103 115 163
Lophoptera 94 101 103 124 168 169
lushanensis 158

Macdunnoughia 117 154 156 161
177 190

Magusa 94 119 167

major 141

mapongua 128

mappa 79 162

megaloba 135

megista 134 149

melanocephala 124

Melanoplusia 190

meretricia 146

metallica 162

micans 124

microgamma 125 133 166 167

mikadina 153

Mimanuga 169

minor 158

minutus 144

mishanensis 151

moneta 150 151

monosigna 156 157

montana 125 133 165 166 167

morigera 150 180

Mouralia 107 141 177 178 180

nadeja 106 155 160
Neoplusia 190

ni 91 96 140 144 145
nichollae 163
Nigramma 127
nigribursa 94 171
nigrisigna 162

nu 154

IAN J. KITCHING

nubeculosa 125
nyiwonis 165 180

obtusisigna 92

ochreata 81 144 145

octoscripta 164

Odontodes 127 169

olivacea 93 134 159
Omorphina 97 103 140 178
Omorphinini 87 140 175
Oncocnemis 94 103 107 108 119
orichalcea 83 92 128 139 145
ornata 151

ornatissima 81 83 93 158
orophila 125 164

ottolenguii 164 180

ou 91 154

oxygramma 85 91 128 134 144 148

Paectes 94 107 169

paghmana 172

Palaeographa 102 103 115 164 165
180

Palaeoplusia 190

palligera 124 153

Panchrysia 151

paraplesia 149

Paraplusia 190

parilis 78 124 125 133 165 166 167

pasiphaeia 161

Penicillaria 169

peponis 134 142

Perloplusia 190

phocea 128 134

photeina 128

Plusia 83 107 115 163 177 190

‘Plusia’ 87 152

Plusidia 115 152

Plusiina 153 159 175

Plusiinae 138

Plusiini 85 87 142 149 175 179

Plusiopalpa 93 101 103 108 113 148
180

Plusiotricha 97 107 113 133 146 147

Podioplusia 190

pokhara 94 157

polycampta 124

Polychrysia 103 106 107 150 152
178 180

polydamia 190

pratti 146

precationis 79 161

psectrocera 78 164

Pseudeva 115 153

Pseudochalcia 152 174

pseudogamma 161

Pseudoplusia 113 143 179

pseudopyropia 190

pterylota 78 164 177

pulchrina 139 162 180

Puriplusia 117 156

purissima 156

purpurigera 153

putnami 163

pyrenaica 165

pyropia 81 124 155

Rachiplusia 107 108 117 121 153
166

rectangula 79 165 166 177

rectilinea 128

rubida 161

rutulifrons 81 124 155

sackenii 125 133 164

sansoni 125 161

schalisema 156

Sclerogenia 107 117 157 177

Scleroplusia 190

selecta 164

semuirosea 145

semivitta 156

separanda 152

serraticornis 90 152

sestertia 92 128

Shensiplusia 94 171

shisa 93 148

sica 78 150

siderifera 152

sigillata 147

signata 81 128 148 149

snowii 125 164

sphinx 180

splendida 152

squammigera 94

stenochrysis 160

Stictoptera 94 103 112 168 169

Stictopterinae 119 167 168

Stigmoplusia 97 107 108 113 149

subaerea 149

subchalybaea 128

surena 165

Syngrapha 115 161 162 164 166
177 180

tancrei 151

tarassota 81 128 147
Targalla 169
tetragona 156
tetrastigma 92
thaumasia 148
thyatiroides 150
Thysanoplusia 107 145
tibetana 125

tinctoides 141
transbaikalensis 165
Trichoplusia 102 107 144
trigemina 140 141
triopus 180

triplasia 139 140 141
tutti 160

u-aureum 165
Unca 190

v-alba 161
v-argenteum 151
vanderweelei 94 157
variabilis 139 150 152
venusta 163 190
vermiculata 134 147

SYSTEMATICS OF THE PLUSIINAE

verruca 91 148 149 vittata 92 128 148 180 Yerongponga 190

viettet 128 145

violacea 164 177 wukongensis 190

virgula 125 154 166 Zonoplusia 107 144 145 178

viridisigma 164 xanthoides 152 zosimi 124 160

261

wae a
diwaraiy (4° mt wynangonres ¢

yee orid 124 182
posta a)
rand vack ey

imcomapicias SP EOE ORE Veoh,
Lu aver TA!

egqverciled iw ~

ianvermicon ¥} a4 [45 ‘ ay)

sien rg emirate 182 im *
infers \3B152

tog TSA bs
jae 1G j
beech) Umi _*
butane 2 We TAD :
ivee@whi 142 & :
iuereerrte 106 LOT ULE SS
| Hy ae ers ———
7 —
é veretuda bd
ed, a : "
124 102 404 iSe le
ines gt ‘te cm
eur TMT
F ty (77 1h
=
y furia Dt iz se?
oy ad
rs 7
é —
ahs
e fants
~~
ea 7’ —
os »
, a)
Sse) 9

-) Pt io Te af,
eorpiaemrtns HS Led | Le 30-4 hy

\ i “ .

eB le} ei

sont teas robs HO oc
Merrie | et f
Odenton ET a — eed

Cera CT VarITS |= Sore :
Geretieess Th 107 HUT I ET
ronment at at ele a eo
a TY
coneaineniedi! SI tid ‘
ye (ie

meena 14 gl) : F

= yer) Cee
”
Pucak Od [A7) pth

* ; wv) tine ae
Ptmeo—wrte ites acm veh ine ast A ’

io F : Se
Femapewe Sy ; =
mltigaes igi 1% 7
a =

Cy ieee Ri
a ee

~~ oe > =

idb jas rant

pailt é Vir es
rer 408 We 1 oe

y) (Red iy 1b

= 10° 1 |

up 9 pee i /

! ag _pare FO
= ite | >
ae
pin a
py cane nt
omnia Gi Le 17s

British Museum (Natural History)

Milkweed butterflies: their cladistics and biology
P.R. Ackery & R. I. Vane-Wright

The Danainae, a subfamily of the Nymphalidae, contains only some 150 species, yet aspects of
their biology have stimulated far more attention than can be justified by species numbers
alone. In recent years, an expansive literature has grown, considering aspects of their
courtship and pre-courtship behaviour, migration, larval hostplan associations, mimicry and
genetics. The popularity of danaines among biologists can certainly be attributed to this
combination, within one small group, of so many of the factors that make butterflies such an
interesting group to study. The obvious need to place this wealth of biological data within an
acceptable systematic framework provided the impetus for this volume.

Started eight years ago within the conventions of evolution by natural selection and
Hennig’s phylogenetic systematics, the book is now largely about natural history (what the
animals have and do, where they live and how they develop) and natural groups — as revealed
by a form of analysis approaching that practised by the new school of ‘transformed cladistics’,
The authors have prepared a handbook that will appeal to a wide range of biologists, from
museum taxonomists to field ecologists.

425 pp, 12 pp colour, 73 b/w plates, line and graphic illustrations, maps, extensive bibliography.
ISBN 0 565 00893 5. 1984. Price £50.

Titles to be published in Volume 54

Studies on the Old World species of Holothrips (Thysanoptera: Phlaeothripidae)
By S. Okajima

Spectacles and Silver Ys: a synthesis of the systematics, cladistics and biology of the Plusiinae
(Lepidoptera: Noctuidae)
By I. J. Kitching

A review of the Solenopsis genus-group and revision of Afrotropical Monomorium Mayr
(Hymenoptera: Formicidae)
By B. Bolton

Photoset by Rowland Phototypesetting Ltd, Bury St Edmunds, Suffolk
Printed in Great Britain by Henry Ling Ltd, Dorchester