Title of Dissertation: 
ABSTRACT 
Biosystematics and the evolution of gall formation in 
hackberry psyllids Pachypsylla (Insecta: Homoptera: 
Psylloidea: Psyllidae) 
Man-Miao Yang, Doctor of Philosophy, 1995 
Dissertation directed by Charles Mitter, Associate Professor, Maryland Center 
for Systematic Entomology, Department of 
Entomology; and Douglass R. Miller, Research 
Entomologist, Systematic Entomology Laboratory, 
Beltsville Agricultural Research Center, USDA 
This dissertation is a study of the phylogeny and evolutionary biology of gall 
formation in psyllids of the subfamily Sponclyliaspiclinae, with particular focus on 
North American hackberry gallers in the genus Pachypsylla. Species in this genus 
produce a variety of gall types on the leaves, petioles, buds and twigs of their 
hosts, four species of Celt is subgen. Euceltis (Ulmaceae ). The homogeneity of 
adult morphology in Pachyp.sylla, contrasted to the great variation in gall 
morphology and phenology, has led to much difficulty in delimiting species. 
Chapter I investigates species limits as related to gall type and host 
specificity in Pachypsylla. Strong differences in al lozymes, morphology and life 
history confirm that leaf, petiole, bud and twig galle rs belong to different species 
or species groups. Different leaf gall morphs probably also represent different 
species, as evidenced by significant allozyme frequ e ncy diffe rences among 
sympatric pairs of gall morphs, consistent frequency difference between co-
occurring morphs across localities, and discrete differences in gall type between 
progenies of individual females . 
Differences in allozymes, female phenology, adult and nymphal coloration, 
as well as laboratory rearings and field manipulations, show that side cell 
individuals within two nipple gall types represent an inquiline sibling species 
(Chapter II). 
Chapter III is an analysis of phylogenetic relationships within Pachypsylla, 
based on allozyme, morphological, life history and chromosome characters. Galler 
populations attacking the same plant tissue form monophyletic groups. The leaf 
galler morphs are little diverged, and phylogenetic relat ionships among them are 
unclear. Populations of inquilines from two different gall types appear closely 
related; the inquiline appears to be derived from a gall-forming ancestor. 
Phylogene tic re la tionships among gallers on different plant parts are consistent 
with an evolutionary sequence of gall position from leaf to petiole to bud to twig. 
Chapter IV is a morphological study of phylogenetic re lationships within 
Sponclyliaspiclinae. The tribe Pachypsyllini, including Pachypsylla and two related 
Celtis feeders, is monophyletic. The tree favors the hypothesis of Burckhardt over 
that of White ancl Hodkinson. The distribution of lerp ancl gall formation is 
shown to be non-random within Sponc.lyliaspidinae. 
Biosystematics and the evolution of 
gall formation in hackberry psyllids Pachypsylla 
(Homoptera: Psylloidea: Psyllidae) 
by 
Man-Miao Yang 
Dissertation submitted to the Faculty of the Graduate School 
of The University of Maryland in partial fulfillment 
of the requirements for the degree of 
Doctor of Philosophy 
1994 
. . ( 
) ., 
. ! 
Advisory Committee: 
Associate Professor Charles Mitter, Chairman, Advisor 
Adjunct Professor Douglass R. Miller, Co-advisor 
Professor Robert F. Denno 
Assistant Professor Charles B. Fenster 
Associate Professor Dr. William 0. Lamp 
li r ,, / 1;//1· / 1 !) ' I ' . (J 
I [__) 
_, / 
,._ - / 
I 
I ! { / ) 
DEDICATION 
To 
Adrienne Venables and Life of Uncertainty 
11 
ACKNOWLEDGEMENTS 
Thanks to N. Moran, C. von Dohlen, R. Patrock, T. Kring, J. Moser, S. 
Heady, C. Trisler, J. Riemann, R. van Devender, D. Riley and B. Farrell for help 
with collecting galls from other states; F. Li, G. T. Xu, L. H. Sun and Y. J. Kwon 
for help with collecting materials in China and Korea; and D. Miller, R. Denno, 
G. Roderick, C. Fenster, W. Lamp and J. Regier for much valuable help and 
advice. Loans of psyllid specimens and constructive comments were provided by 
D. Hollis, National Museum of Natural History, London, D. Burckhardt, Geneva 
Museum, Switzerland, K. L. Taylor, CSIRO, Australia, G. Taylor, University of 
Adelaide, Australia, F. Li, Beijing Agricultural University, China, and Y. J. Kwon, 
Kyungpook National University, South Korea. Special appreciation to G. 
Roderick for the help in data analyses and computer consulting; he and L. Garcia 
de Mendoza also greatly facilitated the electrophoretic work. Helpful guidance 
on SEM technique was provided by J. Plaskowitz, and financial support for SEM 
work was generously provided by D. Miller and M. Stoetzel, Systematic 
Entomology Laboratory, USDA. I also grateful to M. Hardin, T. Jacobs, F. 
Sanders, as well as staff at Catoctin Mountain National Park, Great Falls Park of 
Virginia, State Arboretum of Virginia, Patuxent River Park of Maryland, Palmetto 
State Park of Texas, Boiling Spring State Park of Okalahoma and Beltsville 
Agricultural Research Center, USDA for permission to collect. I especially thank 
A. Mitchell for his help in dissecting galls; S. Cho, G. Hormiga, and A. Wijeskara 
lll 
for helpful discussion and technical support; K. T. Park for his help in 
electropheresis; and T. J. Tsai, G. Li, R. N. Huang, W. Hunter, L. Garcia de 
Mendoza and lab colleagues for help with field work. Without the trucks 
generously provided by many labs in the Department of Entomology, University 
of Maryland, my field work could not be done. I thank the Theodore Roosevelt 
fund of the American Museum of Natural History, the Maryland Agricultural 
Experiment Station, the Smithsonian Institution, the Graduate School of the 
University of Maryland and the USDA NRI-CGP for financial support. Last, but 
not least, I especially thank my family for tremendous supports in various ways, 
C. Mitter for research advising and constant encouragement, and R. Wijeskara, 
L. Garcia de Mendoza, A. Naksathit and S. C. Hung for moral support and many 
wonderful meals during the intense period of my dissertation writing. 
IV 
TABLE OF CONTENTS 
SECTION PAGE 
List of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii 
List of Figures ..... .... ... .. .. .... .. . ... ... . ...... ........... x 
General Introduction and Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 
Chapter I. Species limits as related to gall type and host specificity in 
hackberry psyllids, Pachypsylla .. ........ ..... ....... . ... 8 
Introduction ..... . ..... . ................ .. ......... 8 
Pachypsylla Background ...... . ........ . .... ... ..... .. 9 
Host Plant Celtis Background ................ .. . . ..... 16 
Questions Addressed and Specific Objectives . . . . . . . . . . . . . . 18 
1. Do galls on different plant tissues represent different 
species? ..................................... 19 
2. Are different leaf blade gall types different species? . . . . 19 
3. Do Pachypsylla of similar gall types on different host 
. d'ff . ? 20 species represent I erent species . . ... .. .. .. ...... . 
Material and Methods . .... . .. .. ... . .. . . ....... ...... 20 
Samples . .. ............ .. ..... .. .......... . . . .. 20 
Electrophoretic methods ..... ... ... .. . . . . . . .. .. ... . 23 
Electrophoretic data analyses ....................... 24 
Life history traits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 
R . . eanng expenments ..... ... ...... . . ... ........ ... 25 
Morphological methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 
Results . .. .......... .. . .. .. . .. .. ............... . . 29 
1. Do galls on different parts represent different species? . . 29 
2. Do the different leaf blade gall morphs represent 
different species? . .... . ..... . ..... .. ........ . .. 31 
Allozyme evidence . .. . . . ...... ... ....... . . .. 31 
Life history evidence . .. .... ... ... .. . .. . ..... . 33 
Rearing evidence . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 
Morphological evidence ......... . .. .... .. . ... 35 
3. Do Pachypsylla of similar gall type on different host 
species represent different species? ................ . 35 
Discussion and Conclusions ............ . ....... . ... . .. 36 
Key to Species of Pachypsylla Studied ... .. . . . .... . . ..... 39 
V 
Chapter II. The evolution of inquilinism in hackberry leaf galling psyllids: 
Are multiple individuals within the same gall conspecific? .... 42 
Introduction ...................................... 43 
Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 
Gall inquilines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 
Occurrence of multiple individuals in Pachypsylla leaf galls . 45 
Questions Addressed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 
1. What is the nature of side cell individuals within a leaf 
gall? ....................................... 48 
2. Are side cell individuals from different types of leaf 
galls conspecific? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 
Material and Methods ............................... 49 
Samples for electrophoretic and morphological studies .... 49 
Electrophoretic and morphological methods ... .. .... . .. 50 
Rearing experiments .............................. 51 
Destruction experiment . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 
Female phenology ......... ...... ......... . ... ... 54 
Results ........................ ................. . 54 
Electrophoretic data ..... ....... .................. 54 
Single female rearing experiment ... . ................ 55 
Destruction experiment . . ................ ....... ... 56 
Timing of female appearance ....................... 57 
Morphological comparison ......................... 57 
Discussion and conclusions ................ .... ....... 58 
Chapter III. Phylogeny and evolution of galling position in Pachypsylla . . .. . 62 
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 
Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 
Material and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 
Samples and electrophoretic methods . . . . . . . . . . . . . . . . . 65 
Morphological methods .. ...... ..... ...... ......... 66 
Chromosome data ............................ . .. 67 
Life history data . .............. . .. .. .... . . . ..... . 67 
Phylogenetic analyses ............................. 67 
(1) Electrophoretic data ......................... 67 
(2) Morphological data .................... . .. ... 68 
(3) Combined evidence from morphology, allozyme, 
karyotype and life history . . . . . . . . . . . . . . . . . . . . . 69 
Results ......................... . ................ 70 
Pachypsylla phylogeny .... . ........................ 70 
(1) Trees based on electrophoretic data .. ... ........ 70 
(2) Claclograms based on morphological data ......... 71 
Vl 
(3) Trees based on combined evidencerom morphology, 
allozyme, karyotype and life history . . . . . . . . . . . . . . 72 
Discussion and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 
Monophyly and diversity of gall position groups .......... 73 
Evolution of gall position ................ . ......... 74 
Chapter IV. Phylogenetic relationships of PachypsyIIini in Spondyliaspidinae 
and evolution of lerp and gall formation in Spondyliaspidinae . 76 
Appendices 
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 
Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 
Systematics of Pachypsyllini . . . . . . . . . . . . . . . . . . . . . . . . . 78 
Systematics of Spondyliaspidinae . . . . . . . . . . . . . . . . . . . . . 82 
The position of Pachypsyllini within Spondyliaspidinae .... 83 
Conceal-feeding (Ierp- and gall-forming) habits in 
Spondyliaspidinae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 
Material and Methods .................... . .......... 86 
Sampling of taxa ........... . ..... .. ... . .......... 86 
Morphological methods ... . ................... . . . .. 88 
Morphological data analyses . . . . . . . . . . . . . . . . . . . . . . . . 89 
Life history information, character coding and hypotheses 
testing . .. ................................ . .. 91 
Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 
The phylogenetic relationships ..... . .. . ............. 93 
Lerp and gaII forming biology in Spondyliaspidinae . . . . . . . 94 
The evolution of concealed feeding in psyllids ........... 97 
Discussion and Conclussions . . . . . . . . . . . . . . . . . . . . . . . . . . 98 
Phylogenetic relationships ........ . ...... . . . .. .. . ... 98 
The evolution of concealed feeding in psyllids . . . . . . . . . . 102 
107 
Tables and Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189 
Vll 
LIST OF TABLES 
Numbers Page 
1-1. Species entities described in Pachypsylla (Homoptera: 
Psylloidea). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 
1-2. Synopsis of hackberry psyllid gall types ................... 114 
1-3. Collecting sites of Pachypsylla samples . . . . . . . . . . . . . . . . . . . 116 
1-4. Enzyme loci scored for phylogenetic analysis of Pachypsylla ... 118 
1-5. Allele frequencies of different gall types from 38 populations 
based on adult data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 
1-6. Allele frequencies of different gall types from 22 populations 
( after pooling) based on nymphal data . . . . . . . . . . . . . . . . . . . 123 
1-7. X2 test of allele frequencies among different populations of 
bud and twig gall makers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 
1-8. Body size measurement of Pachypsylla gall makers from 
various gall types. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126 
1-9. X2 tests for allele frequency differences among populations 
of leaf gall makers, all loci . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 
1-10. Frequency of major alleles in IDH-1 in blister gall versus hairy 
nipple gall from same populations . . . . . . . . . . . . . . . . . . . . . . 128 
1-11. Types of gall found in individual female rearing experiment . .. 129 
1-12. Head/shaft ratio of terminal segment of male genitalia of bliste r 
gallers from different hosts . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130 
2-1. Allele frequencies of individuals from different cell position 
of a gall of various leaf gall types . . . . . . . . . . . . . . . . . . . . . . . 131 
2-2. X2 test of allele frequency of gall types from same cell 
position .......................................... 133 
? .., 
--.). All ele frequencies of individuals from different cell positions .. 134 
Vlll 
2-4. X2 test of allele frequency of different cell position . . . . . . . . . . 136 
2-5. Results of single female rearing experiments . . . . . . . . . . . . . . 137 
3-1. Character matrix: morphological characters (1-33); al lozyme 
data (34-48); karyotype (49-50); life history (51-51) coded 
from 1-6. Each locus is treated as single character . . . . . . . . . . 138 
3-2. Characters matrix for all evidences: morphological characters 
(1-33); allozyme data (34-130); karyotype (131-132); life 
history (133-135) coded from 1-6. Each allele is treated as 
single character ...... ........... ............. ... . .. 141 
3-3. Characters used for Pachypsylla phylogenetic study ... .. ..... 146 
4-1. List of genera of Spondyliaspidinae from White & Hodkinson 
versus Burckhardt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150 
4-2. Characters used for phylogenetic analyses of Pachypsyllini and 
Spondylidaspidinae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 
4-3. Character matrix used in Spondyliaspidinae phylogenetic 
analyses based on morphological characters . . . . . . . . . . . . . . . 157 
4-4. Summery of concealment types and character states scored 
for each taxa for analyses of evolution of gall and Ierp 
formation in Spondyliaspidinae . . . . . . . . . . . . . . . . . . . . . . . . . 161 
lX 
Number 
1-1. 
1-2. 
1-3. 
1-4. 
1-5. 
1-6. 
1-7. 
2-l. 
2-2. 
2-3. 
3-1. 
-, ? 
.)-__ 
-, ,..., 
.)-.). 
3-4. 
LIST OF FIGURES 
Drawings of the various types of galls of Pachypsylla 162 
Geographical distribution of host plants, Celtis, of Pachypsylla . . 163 
Map of sampling localities of Pachypsylla adults . . . . . . . . . . . . 164 
Map of sampling localities of Pachypsylla nymphs . . . . . . . . . . . 165 
Forewing pattern variations of the major galling groups of 
Pachypsylla adults ............ . ... .. ................ . 166 
Life history of two leaf gall types in National Agriculatural 
Library population in Maryland . . . . . . . . . . . . . . . . . . . . . . . . 167 
Scanning electron rnicrographs of male proctiger in leaf 
blister galls (a) and other leaf galls (b-d) ................. 168 
Distance Wagner and UPGMA trees, based on Rogers' 
distance (15 loci), of eight populations of Pachypsylla leaf 
gall nymphs from different gall types and cell posi tions . . . . . . . 169 
Change of gall size through time after experimental 
destruction of center cell nymph . . . . . . . . . . . . . . . . . . . . . . . . 170 
Relative frequencies of center versus side cell females 
caught at Beltsville, MD, 1993 ......................... 171 
Femlae genitalia of Pachypsylla, Tetragonocephala and 
Celtisaspis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 
Distance Wagner tree of Pachypsylla based on Roge r' 
distance and rooted with Celtisaspis beijingana, using 
allozyme data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 
UPGMA tree of Pachypsylla with outgroups Tetrogonocephala 
/lava and Celtisaspis beijingana based on (a) Nei's and (b) 
Roger' distance using allozyme data ..................... 174 
Strict consensus tree from 3400 most parsimonious trees from 
allozyme data, loci treated as characters, alleles as character 
X 
states ............................................ 176 
3-5. Strict consensus tree from four most parsimonious trees from 
allozyme data with alleles treated as separate characters . . . . . . 177 
3-6. Three equaly parsimoniou trees (a-c) and their strict consensus 
tree ( d) of Pachypsylla, unordered morphological data . . . . . . . 178 
3-7. Three equaly parsimonious trees (a-c) and their strict consensus 
tree ( d) of Pachypsylla, ordered morphological characters 179 
3-8. Scanning electron rnicrographs of Pachypsylla adult heads 180 
3-9. Consensus trees for combined data ................ .. .... 181 
3-10. Caudal spurs of the leaf blister gall maker. (a) dorsal view 
of 5th ins tar nymph; (b) enlargement of caudal spurs, do rsa l 
view; (c) enlargemnet of caudal spurs, ventra l view . .. .. .. ... 182 
3-11 . Hind coxa of hackberry leaf star gall maker showing: (a) distal 
ventral crescentic denticulate patch and a fan-sh aped area near 
the patch, ventral view; (b) lateral view; (c) enlargement of 
denticulate patch of (a); ( d) enlargement of hole between fan-
shaped area and trochanter . . . . . . . . . . . . . . . . . . . . . . . . . . . 183 
4-1. Partial phylogeny of Psylloidea after White and Hodkinson 
(1985). Highlighted taxa indicate those sampled for current 
study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184 
4-2. Types of occipital foramen ....................... . .. . . 185 
4-3. Three equally parsimous trees (a-c) and their strict consensus 
( d) for Spondyliaspidini based on morphological data with all 
characters unordered. Same trees found after successive 
,ve1ght1ng . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186 
4-4. Strict consensus of 164 most parsimonious trees for 
Spondyliaspidini based on morphological data with so me 
characters (#3-7) ordered and others unordered ............ 187 
4-5. Strict consensus of 10 most parsimonious trees for 
Spondyliaspidini after successive weighting with some 
characters (#3-7) ordered and others unordered . .. . ....... . 188 
Xl 
GENERAL INTRODUCTION AND OBJECTIVES 
The biology of gall forming insects represents one of the most conspicuous 
and complicated animal-plant interactions known. Galls have been utilized 
extensively in medicine, industry, and food for over a thousand years in ancient 
countries such as India and China. The first scientific report on insect-induced 
galls was published in the seventeenth century by Malpighi in Italy (Mani, 1992), 
but most subsequent reports were based on casual observation. In recent decades, 
however, serious and intensive studies have been undertaken on numerous aspects 
of galling, including gall morphology and development, physiology and 
biochemistry of gall tissues, life history, behavior, host specificity and nutritional 
physiology of gall makers, and also trophic relationships of the complex 
assemblages of organisms associated with galls (Mani, 1992). Contemporary 
studies report gall making from seven insect orders with their biology proving 
remarkably diverse. However, there is still little understanding of the evolutionary 
origin, adaptive significance, and ecological and evolutionary consequences of gall 
formation. 
This dissertation is a study of the phylogeny and evolutionary biology of gall 
formation in psyllids of the subfamily Spondyliaspidinae, with particular focus on 
North American hackberry gallers in the genus Pachypsylla. 
The Psylloidea (Insecta: Homoptera: Sternorrhyncha) comprises about 2000 
species in the world (Hodkinson, 1984 ). Psyllids feed on a wide range of di cots, 
1 
with a few species on monocots (Juncus, Carex and the palm Pritchardia) and 
conifers (Pinus). They are usually narrowly host specific, and species that initially 
appear to feed on more than one plant family are likely to prove to represent 
complexes of host specific sibling species (Hodkinson, 1986; Klimaszewski, 1964; 
Yang et. al, 1986). 
Psyllids are sucking insects, feeding primarily on the soluble contents of 
phloem tissue, though some species attack mesophyll tissues (Woodburn and 
Lewis, 1973; Hodkinson, 1973, 1984; White, 1970). In Roskam's view (1992), 
feeding on plant sap offers homopterans extensive opportunities for manipulating 
host plants, accounting for the repeated evolution of gall formers in this group. 
Gall-forming psyllids are broadly distributed across plant taxa and 
geographical regions. There are at least 350 gall-inducing species (Mani, 1964; 
Hodkinson, 1984; Dreger-Jauffret and Shorthouse, 1992), constituting more than 
15% of all psyllid species. Psyllid gall forms range from simple distortion of 
plants, such as leaf curling and leaf pit galls, to a highly complex structure, such 
as the sealed gall of Schedotrioza or that of Trioza magonoliae, which develops a 
dehiscent mechanism when mature (Ashmead, 1881; Crawford, 1914; Morgan, 
1984 ). Psyllid galls usually show a high degree of site specificity on their host 
plant. 
The complexity and specificity of ecological relationships in galling psyllids 
are well illustrated by the North American genus Pachypsylla, commonly known 
as hackberry psyllids. Species in this genus produce a variety of gall types on the 
2 
leaves, petioles, buds and twigs of their hosts (Fig.1-1), four species of Ce/tis 
subgen. Euceltis (Ulmaceae), whose geographic distributions are partially 
overlapping (Fig. 1-2). The homogeneity of adult morphology in Pachypsylla, 
contrasted to the great variation in gall morphology and phenology, has led to 
much difficulty in delimiting species. Over two dozen specific entities have been 
named, but the status of most of these is entirely unclear. 
Goals 
The first goal of this project is to combine electrophoretic, morphological 
and life history data to determine species limits and phylogenetic relationships in 
Pachypsylla, as a basis for understanding speciation and the evolution of host use 
and gall formation in this group. The second objective is an analysis of 
phylogenetic relationships in the subfamily Spondyliaspidinae, with particular focus 
on the monophyly and phylogenetic position of Pachypsylla. The third objective 
is a comparative analysis of galling and related habits in the subfamily, asking 
whether several evolutionary patterns suggested by Pachypsylla and allied genera 
reflect broader trends. The literature on spondyliaspidine life history is 
synthesized, and the distribution of galling and related habits mapped on the 
morphological phylogeny. 
Significance 
Over the long term, these and my intended subsequent studies will bear on 
3 
several general questions about the nature of evolution in gall formers. The first 
concerns the role of ecological specialization in speciation. Perhaps more than 
any other phytophagous insect, gallers meet the broad definition of "parasites" 
advanced by Price (1980). There are several reasons to suppose that ecologically 
highly specialized organisms, exemplified by parasites, should have unusually high 
rates of diversification (Price, 1980; Futuyma and Moreno, 1988). Parasites, 
including galling psyllids, are typically modified and intimately dependent for 
survival on only one or a few types of hosts or even parts thereof, which often 
constitute both food and habitat for the much smaller parasite. They should thus 
be highly subject to diversifying selection arising from variation within and among 
host species. Reproductive isolation could result from such selection either 
directly, e.g. through adaptive divergence in phenology (Wood, 1980; Wood and 
Guttman, 1982; Wood et. al. 1990; Bush, 1969; Prokopy et. al. 1988), or indirectly, 
through pleiotropy (Maynard Smith, 1966; Ringo, 1977; Futuyma, 1986). 
Ecological specialization could also promote diversification by reducing 
competition between incipient species; this is the gist of Mayr's assertion (1976) 
that "extreme specialization is characteristic in insects and explains their 
prodigious rate of speciation." 
A necessary first step in assessing the role of host adaptation in the 
speciation of galling psyllids is documentation of the degree and kinds of 
differences in host use between closely related species. Do nearest relatives differ 
most prominently in host species, in plant part attacked, or in phenology with 
4 
respect to that of the host; or do they instead show great overlap, suggesting that 
niche specificity is not crucial to speciation? This study of species limits in 
Pachypsylla is one of the first to explicitly address this question for galling insects. 
The "specialization-diversification" hypothesis above also makes two 
broader predictions, which I hope to address in a long-term phylogenetic analysis 
of psyllids as a whole, for which this project is a first step. First, gall-forming 
species should show greater specialization in host taxon use than related non-
gallers. Second, clades of gallers should be consistently more diverse than closely-
related non-galling clades of the same age, i.e. their sister groups. 
In strong contrast to the "specialization-diversification" hypothesis as 
applied to "parasites" in the broad sense, an older view holds that ecological 
specialization particularly as exhibited by parasites is a "dead end" sharply limiting 
the potential for subsequent evolutionary change or diversification (Futuyma and 
Moreno, 1988; Moran, 1988). On a broad scale, this hypothesis predicts that 
evolutionary reversion to free living from galling should be rare, gall formers may 
shift among host taxa less often than related free-living forms, and gall formers 
should be less diverse than their non-galling sister groups. 
If galling imposes strong constraints on niche shifts, we might also expect 
that evolutionary transitions to different types of galling habit would occur in small 
steps (Janson, 1992). This should be reflected in phylogenetic sequences 
predictable from the probable degree of genetic divergence between habits. For 
example, it could be predicted that in Pachypsylla, leaf blade galling should 
5 
represent the ancestral gall position, as leaf feeding is typical of psyllids as a 
whole; attack of the petiole, bud and twig may represent successively more 
different and evolutionarily more recent habits. A similar sequence was 
postulated for leaf, bud and twig gallers in the sawfly subfamily Nematinae (Smith, 
1970; Price, 1988, 1992). 
On a broader scale, it has been hypothesized that the origin of true gall 
formation itself may occur in a predictable sequence from other forms of 
concealed feeding. Thus, construction of a lerp ( a cover constructed from 
hardened honeydew and probably mixed with wax; see Chapter IV) and of partial 
galls by related species in related genera have been suggested by Hodkinson 
(1984) as precursors to the enclosed galls formed by Pachypsylla. 
My phylogenetic analyses test these postulates for Pachypsylla and relatives. 
The comparative study across Spondyliaspidinae and related genera permit a 
preliminary statistical testing for the generality of such trends, as all of these 
habits have probably arisen multiple times. 
This dissertation is divided into four chapters, each will be a separate 
publication. The first chapter treats the analyses of species limits within 
Pachypsylla. The second chapter examines the occurrence of multiple individuals 
within leaf galls, and provides the first demonstration of a gall inquline in psyllids, 
and discusses the evolution of inquilinism in Pachypsylla. The third chapter is a 
phylogenetic analyses of relationships and the evolution of gall types within 
6 
Pachypsylla and the two other genera in Pachypsyllini, based on allozyme and 
morphological data. The fourth chapter examines phylogeny in the subfamily 
Spondyliaspidinae, test hypotheses of previous authors including White and 
Hodkinson (1985), then uses the phylogeny estimate to examine the evolution of 
galling and lerp formation. 
7 
Chapter I. 
Species limits as related to gall form and host specificity 
in hackberry psyllids, Pachypsy/la 
8 
INTRODUCTION 
Pachypsylla is one of the most conspicuous gall formers in North America, 
producing a variety of gall types on the leaves, petioles, buds and twigs of its hosts, 
Ce/tis subgen. Euceltis (Ulmaceae) (Fig. 1-1 ). The gall types are differentially 
distributed across four species of Celtis whose geographic distributions are partially 
overlapping (Fig. 1-2). Within the gall makers in the same position in the tree, 
adult morphology is quite homogeneous but there is great variation in gall 
morphology, especially among the leaf gall makers. This has led to widely 
differing views on the number of species in Pachypsylla. 
This study tests several hypotheses on species limits of Pachypsylla as 
related to galling position, gall shape variation, and host specificity, as a pre-
requisite for understanding speciation in this group. 
PACHYPSYLLA BACKGROUND 
The first described species of what is now considered to be Pachypsylla was 
the petiole gall maker, Psylla venusta Osten Sacken. In a short paragraph, Osten 
Sacken (1861) delineated the shape and seasonal change in texture of petiole galls 
and mentioned the large body size and dark maculated wings of the adults. 
Although he treated the insect as Psylla venusta, he pointed out that peculiarities 
of venation and the form of the metasternal spur indicated that it should be 
placed in a new genus. Riley first described the genus in 1883 and gave a short 
9 
description of the gall and nymph. He provided a more extensive description two 
years later (Riley, 1885) for the genus and three species, P. venusta, P. celtidis-
mamma and P. (Blastophysa) celtidis-gemma. 
Twenty-four names (Table 1-1) have been used for species of Pachypsylla, 
based on gall structure, host differences, and adult morphology. The status of 
many of these entities is uncertain. Of the two recent treatments, Tuthill (1943), 
who followed Crawford's system (1914), recognized 7 species, while Riemann 
(1961) in his unpublished dissertation recognized 12 (Table 1-1). 
Members of Pachypsylla can be divided into four groups based on nymphal 
and adult morphology and on the position of the galls on the hackberry host. The 
groups are the petiole gall maker, Pachypsylla venusta, leaf-blade gall makers (here 
after termed simply leaf gall makers), Pachypsylla spp., the bud gall maker, 
Pachypsylla celtidisgemma, P. pallida & P. dubia, and the twig gall maker, P. 
celtidisinteneris. 
It is generally accepted that there is only one species of petiole gall maker. 
Riley (1883) synonyrnized P. celtidisgrandis, which he had described as a new 
species in his earlier publication (Riley, 1876), with P. venusta, and Crawford 
(1914) synonyrnized P. tridentata (Patch, 1912) with P. venusta; their actions left 
only one eligible name for the petiole gall maker. 
Bud and twig gall makers are similar in morphology. Several species of 
bud and twig gall makers have been named, but these are mostly based on little 
information. Tuthill (1943) recognized four of these, namely P. celtidisgemma, P. 
10 
celtidisinteneris, P. dubia and P. pallida. The first two species, P. celtidisgemma and 
P. celtidisinteneris, are bud and twig gall makers, respectively. The latter two 
species were described by Patch (1912) based on specimens in the Cornell 
collection. They are similar morphologically to the bud gall maker P. 
celtidisgemma, except in forewing pattern. P. celtidisgemma has uniformly 
immaculate wings, whereas the wings of P. dubia are densely mottled and those 
of P. pallida are shaded, with a pale streak extending transversely across the 
forewing. Patch associated only the specimens of P. dubia with a Celtis gall, while 
no gall data were indicated for P. pallida. Tuthill (1943) identified specimens 
accompanied by galls from several localities in Arizona and New Mexico as P. 
pallida. These are very densely pubescent bud galls. The subgenus Blastophysa 
was erected by Riley (1885) for P. celtidisgemma but was not accepted by most 
later workers. 
The taxonomic situation is most confusing within the leaf gall making 
complex. At least twelve names have been assigned to the leaf gall makers. Riley 
looked at gall variations and named many new species based more on the gall 
rather than on the insect itself in his series of publications (1876-1890) on 
hackberry galls. He provided a key to three species of Pachypsylla based mainly 
on adult morphology, but gave a key and description to the galls for 9 species and 
a variety, including 6 new species (Riley, 1890). Mally (1894) studied the biology 
of hackberry psyllids at Ames, Iowa. Unlike Riley, Mally individually separated 
galls of variant nipple shapes in pill boxes and reared the adults. Because there 
11 
was great variation in adult characters but no constant association of these 
variants with gall shape and size, he concluded that they were all part of a single 
variable species. A year later, however, Mally (1895) listed several names that 
Riley described, such as P. celtidis-cucurbita, and P. celtidis-pubescens which form 
different nipple-shaped galls. It is not clear to what extent Mally considered gall 
variation to be species specific. 
Crawford (1914) adopted Mally's skepticism about Riley's species and 
considered all leaf gallers except the blister gall maker to represent P. 
celtidismamma. From then on, many authors looked at gall shape and/or adult 
morphology and named new species or synonyrnized old species (e.g. Caldwell, 
1938; Tuthill, 1943). Tuthill (1943) in his revision of North American psyllids 
separated the leaf galling species by body size; he recognized two species, the 
blister gall maker, P. celtidisvesicula (2.5mm or less), and the nipple gall maker, 
P. celtidismamma (3 to 4 mm). 
Riemann (1961) reported that there is no variation in chromosome number 
among leaf gall makers including P. celtidisasterisca, P. celtidispubescens, P. 
celtidismamma, P. celtidisvesicula and five undetermined species, although there 
is variation among gallers on different plant tissues. Various species of 
Pachypsylla leaf gall makers and the twig galler, P. celtidisinteneris, had the same 
chromosome numbers (2N = 25 d and 26 9) as did their closest North American 
relative, Tetragonocephala flava. Both the petiole galler, P. venusta (2N = 23 d , 
24 9 ), and the bud galler, P. celtidisgemma (2N = 22 d, 22 9 ), had smaller 
12 
chromosome numbers. All of these were reported to have an XO sex 
determination system except P. celtidisgemma which was described as having XY. 
Riemann's findings, however, did not entirely agree with those of Walton 
(1960). Four species of Pachypsylla from New York were studied by Walton, 
including three leaf gall makers and a bud galler. Unlike Riemann, Walton found 
variation between leaf gallers; both P. celtidismamma and P. celtidisvesicula were 
reported to have a 2N chromosomal number of 27 in males and 28 in females 
while P. celtidiscucurbita had 2N = 25d, 269. Like Riemann, Walton found fewer 
chromosomes in the bud gall maker (2N = 2ld, 229) than leaf gallers, but he 
reported a different male chromosome number. Chromosome number variation 
in Pachypsylla remains unclear. 
Gallers on different host tissues also differ in which stage overwinters. The 
leaf gall makers emerge from the gall just before leaf drop in the fall and 
overwinter as adults. Where the leaf gall makers overwinter is uncertain. Mally 
(1894) found leaf gall makers in the cracks and crevices of hackberry bark in 
March 1892 while the weather was still cold in Iowa. Some authors (e.g. Tuthill, 
1943) reported that psyllids were found in large numbers on screens and 
frequently became a nuisance in the house in autumn. Most workers (e.g. Wells, 
1920; Tuthill, 1943) have assumed, without real evidence, that these psyllids 
overwinter in bark crevices or leaf litter. 
The petiole, bud and twig gallers overwinter inside the gall on the tree in 
the last nymphal instar and emerge the following spring. All Pachypsylla nymphs 
13 
j 
emerge by sawing their way out of the gall using the spiny chitinized tip of the 
abdomen. 
All species of Pachypsylla mate and lay eggs in the spring. However, there 
are phenological differences among gall types. The petiole gall makers emerge 
from the gall in very early spring (mid-March to early April in Maryland) as soon 
as the weather turns warm. If petiole galls are brought into a warm room in the 
winter, nymphs emerge from the gall and molt to adults several hours later. The 
leaf gall makers come back to the tree in early spring when the buds begin to 
swell and are present until the leaves are fully extended (from late March/early 
April to late May in Maryland). They usually mate and lay eggs on the bud scale 
or underside of the leaf. In late spring (late May to early June in Maryland), 
when the year's new twigs are formed but still green, the bud and twig gall makers 
emerge, mate and lay eggs. 
Differential times of occurrence were also observed between different types 
of leaf gall makers in the National Agricultural Library population. Of the two 
types of leaf galls which co-exist on this tree, even on the same leaf, nipple galls 
were formed earlier than star galls. 
The petiole galls are always polythalamous or multi-chambered, each 
chamber confining a single nymph. The bud galls are usually polythalamous but 
sometimes there is only one individual inside a gall whereas a twig gall always 
harbors only one individual. Both single and multiple-cell galls are found in most 
leaf galls except blister galls on C. occidentalis. The species status of the 
14 
individuals within one gall is discussed in Chapter IL 
A final uncertainty concerning species limits in the leaf gallers is the status 
of populations on different host species. Some species of Pachypsylla use several 
hosts and are widely distributed over the range of subgenus Euceltis in the United 
States, e.g. P. venusta and P. celtidisinteneris. Others are largely restricted to 
particular species of Celtis (Table 1-2). For example, in the northeastern United 
States, the hairy nipple gall (P. celtidismamma) and the blister gall (P. 
celtidisvesicula) are commonly associated with C. occidentalis, while the glabrous 
nipple gall (probably P. celtidisglobulus) and the star gall (P. celtidisasteriscus) are 
usually on C. tenuifolia. 
Riemann (1961) postulated strict host specificity within two of the leaf gall 
making groups, the nipple gall group and blister gall group. He named two new 
species (Table 1-2, new species 1 and 2) comprising the blister gall makers feeding 
on Celtis laevigata and C. reticulata, respectively, and restricted the name P. 
celtidisvesicula to blister gallers on C. occidentalis. Similarly, he restricted P. 
celtidismamma to the occidentalis nipple gall maker, applying the name P. 
celtidispubescens to nipple gall makers on C. reticulata, and P. sp. (probably P. 
celtidiscucurbita var. "?" Riley) to those on C. laevigata. Members of the nipple 
gall group were distinguished by host and gall shape differences as well as slight 
dissimilarities of male genitalia. The blister gallers were distinguished by host 
differences and slight differences in the male genitalia, mainly the head/shaft ratio 
of the terminal segment of the aedeagus. 
15 
Riemann attributed some exceptions to host specificity to hybridization 
among Celtis species. Hybridization and introgression among hosts are not rare. 
The species of Euceltis are very similar to each other, and there has been difficulty 
in defining species limits (Elias, 1970; Wagner, 1974). 
HOST PLANT CELTIS BACKGROUND 
Celtis belongs to the elm family, Ulmaceae. It consists of about 70-80 
species, found in both temperate and tropical regions (Rehder, 1949; Preston, 
1961; Correll and Johnston, 1970; Kriissmann, 1976). 
Celtis has been used in the manufacture of furniture, as an ornamental 
plant, as a shade tree in Temperate region (Hough, 1936) and as material for 
making ropes and papers in Asia. The native American Indians used the fruits as 
a flavoring for meat or finely pounded the berries and mixed them with parched 
corn (Gilmore, 1919; p.76). The common hackberry, C. occidentalis, has many 
desirable traits as a shade tree, such as an upright crown with lower branches 8 
to 10 feet from the ground, and considerable drought hardiness (Albertson and 
Weaver, 1945). It is receiving attention in the Great Plains because of the disease 
problems that have developed in several tall-tree species such as elm (Anderson 
and Tauer, 1993). 
Planchon (1873) divided Celtis into four subgenera on the basis of floral 
morphology and geographic distribution. Two of these occur in North America, 
i. e. Euceltis and Momesia. Euceltis is distributed in temperate regions of the 
16 
northern hemisphere and high mountains of the tropics. Momesia is found in 
tropical and subtropical parts of the western hemisphere. 
Five species of Celtis are found in North America. The desert hackberry, 
C. pallida, is a member of the subgenus Momesia. The remaining four species 
belong to the subgenusEuceltis, namely the common hackberry, Celtis occidentalis, 
the dwarf hackberry, C. tenuifolia, the sugar hackberry, C. Zaevigata, and the 
netleaf hackberry, C. reticulata. They support as many as 17 or even more types 
of insect galls, though these pests seldom kill the tree (Elias, 1970; Wells, 1916). 
Arthropod galls are so common on hackberries that some field botanists use these 
galls, e.g. witches brooms caused by mites and petiole galls of hackberry psyllids, 
as a way to identify these deciduous trees in the winter (Wagner, 1974). 
Pachypsylla and the other two genera of the tribe Pachypsyllini feed only on 
species of Euceltis. Apart from Pachypsylla, the monotypic North American genus 
Tetragonocephala is known from C. reticulata and C. Zaevigata, while Celtisaspis, 
found in China, Japan and Korea, has been reported from Celtis sinensis and C. 
bungeana, two of the approximately 26 species of subgenus Euceltis in Asia 
(Rehder, 1949). 
Of the host plants of Pachypsylla, Celtis occidentalis has the widest 
distribution (Fig. 1-2), occurring mainly in mid-western and north-eastern North 
America. It is commonly found in rich, moist soil along stream banks or on flood 
plains (Stephens, 1973) usually in semi-shade areas. Celtis tenuifolia has a patchy 
distribution in the eastern region. It prefers limestone regions, growing upon 
17 
rocky bluffs, in rocky woods, and in open glades (Stephens, 1973; Wagner, 1974). 
Some authors consider C. tenuifolia to be a subspecies or variety of C. occidentalis 
(e.g. Pepoon, 1927), but most recent researchers have recognized it as a true 
species (e.g. Kriissmann, 1976). Wagner (1974) compared the two species in the 
Great Lakes region and confirmed the species distinctness of C. tenuifolia. He 
reported that in addition to its "normal" range, which is well below the line of 
maximum Wisconsin glaciation, this species occupies scattered localities in the 
Great Lakes region. I also found the distribution of this species to be far broader 
than depicted by Little (1971, 1976, 1977) (Fig. 1-2). Many populations were 
found in Maryland, Virginia and Delaware, sometimes closely co-occurring with 
C. occidentalis (e.g. at Great Falls, Virginia). The sugar hackberry, Celtis laevigata, 
is a southeastern species while C. reticulata is a southwestern species. The 
distributions of the two species overlap in central Texas and Oklahoma. 
Species of Euceltis are very difficult to distinguish, because there are many 
varieties and natural hybrids (Rehder, 1949; Preston, 1961; Stephens, 1973; 
Kriissmann, 1976). Named varieties tend to be intermediate between species. For 
example, C. laevigata var. taxana seems to be intermediate between C. reticulata 
and C. laevigata (Sargent, 1922). Thus the taxonomy of Celtis is in a confused 
state and needs revision. 
QUESTIONS ADDRESSED and SPECIFIC OBJECTIVES 
Complete revision of the Pachypsylla species is one long-term aim of my 
18 
work, but this goal will take many years. My thesis work has focused on several 
hypotheses about the nature of species differences in this genus. These include 
contrasts among gallers on different host tissues, on different host species, and 
with different gall morphologies and phenologies. 
The following specific questions are addressed. 
1. Do galls on different plant tissues represent different species? 
Nymphs dissected from and adults emerging from leaf, bud, petiole, and 
twig galls were compared by electrophoresis, to check previous conclusions that 
they are distinct species, and to gauge the power of allozyme data for separating 
species in Pachypsylla. 
2. Are different leaf gall types different species? 
The following hypotheses are contrasted: 
A. The null hypothesis is that all the leaf gall makers, despite making different 
gall types, are a single species. 
B. Under the "two species hypothesis" of Crawford (1913) and Tuthill (1943), 
only the blister galler (P. celtidisvesicula) is distinct from all other gallers 
(P. celtidismamma). 
C. Under the "multiple species hypothesis" of Riley (1876- 1890) and Riemann 
(1961), there are multiple species (up to seven) of leaf gall makers 
distinguished by gall shape. 
Different types of leaf gallers from different hosts across a wide 
19 
geographical area were sampled and analyzed morphologically and electro-
phoretically. Possible differences in life history were examined, and rearing 
experiments were conducted to test whether the offspring of an individual female 
all produce the same gall type, as expected if gall shape is genetically determined 
and species specific, rather than being determined by, for example, leaf 
developmental stage. 
3. Do Pachypsylla of similar gall type on different host species represent different 
species? 
The hypothesis of Riemann (1961), who proposed that blister gall makers 
on different host species are different species, was tested. Sympatric samples of 
nymphs and/or adults from Celtis laevigata and C. reticulata were compared using 
morphological and allozyme analysis. 
MATERIAL AND METHODS 
Samples: 
Extensive collections of adults from sixteen localities across the United 
States were obtained in the fall of 1991. Galls were collected by myself and lab 
colleagues in the greater Washington area and by cooperators in Ohio, Arkansas, 
Louisiana, Texas and Arizona (Table 1-3 and Fig. 1-3). Galls were collected in 
the fall before the adults emerged. I sorted the galls by shape into different bags, 
collected emergent adults from the bags, and froze them at -80°C. Seven gall 
types, including 3 populations of petiole gall makers, 3 populations of bud gall 
20 
makers and 32 populations of leaf gall makers were included in this survey. These 
samples represented all five leaf gall types. Owing to its rarity and the high 
parasitism rate in the greater Washington area, the twig gall was not included in 
this survey of adult. Each type of gall on each host in each locality was treated 
initially as a separate population. Samples of each population were collected from 
1-3 trees. Sample sizes for individual populations ranged from two to 33, with 
most being 8 or more. For pairwise comparison of 5 sympatric leaf gall types at 
Great Falls Virginia (GFV), three (blister gall, hairy nipple gall and glabrous 
nipple galls) of the five gall types were taken from the same tree. The other two 
gall types were collected from two other separate trees. All these trees are C. 
occidentalis and are located within 0.5 km of each other. 
Following the discovery that adults emerging from leaf galls with multiple 
cells may include both the gall former and inquilines (see Chapter 11), a second 
set of samples, of nymphs, was obtained in the fall of 1992. These individuals 
were collected by dissecting individual galls from multiple branches of one to five 
trees. Only individuals from mono-cell galls or from the center cell of multiple 
cell galls were used to represent each gall type. The sample sizes were 14 to 46 
individuals for each population for leaf gall makers, and 4 to 13 per population 
for other gall makers. Twenty-four populations were examined, including three 
populations of the petiole gall maker, four populations of glabrous bud gall 
makers, one population of hairy bud gall maker, one population of twig gall maker 
and 15 populations, including all five types, of the leaf gall makers (Table 1-3, Fig. 
21 
1-4). 
For pairwise comparisons of sympatric gall types, three populations with 
multiple gall types were studied. Three gall types, blister gall, disc gall and 
glabrous nipple gall, were collected from two trees, both bearing all three gall 
types in the GFV population. These two trees are ten meters from each other in 
an open area. The two gall types, blister and hairy nipple gall, from Catoctin 
Mountain (CMt), Maryland, were from four and three trees, respectively, with two 
of these trees being sampled for both gall types. All the trees are in the Poplar 
Grove area within 0.5 km of each other. Both glabrous nipple gall and star gall 
were found on the same tree in the National Agricultural Library (NAL) and 
Branchville Road, Berwyn (BB) populations in Maryland. Samples of these are 
based on a single tree from each locality. The NAL and BB populations are 
about 5 km apart. 
For testing Riemann's hypothesis on host specificity, I made a trip to the 
Southwest, including Riemann's original localities, in the fall of 1992. Riemann 
(1961) regarded populations of blister gall makers on Celtis laevigata and C. 
reticulata in the southwestern U.S. as separate species. My collecting sites 
included the following Texas localities: Brackenridge field lab of the University 
of Texas at Austin; Zilker Park, Austin; and Palmetto State Park, Gonzales. In 
each locality, samples were collected, dissected and frozen from two or three sets 
of paired trees of C. laevigata and C. reticulata, located from O to 10 meters apart 
with no intervening barriers. 
22 
Electrophoretic methods: 
Allozyme electrophoresis was carried out using cellulose acetate gels, 
following the methods of Hebert and Beaton (1989) and Easteal and Boussy 
(1987). The Titan III cellulose acetate gel apparatus from Helena Laboratories 
was used, with gels measuring 94 x 76 mm. This allowed twelve samples to be run 
side-by-side at the same time. 
Prior to sample preparation, body sizes of each individual was measured 
using electronic caliper. Frozen individual psyllids were homogenized in 5 µI of 
distilled water in the sample well, using a glass rod cut from microscope slides. 
Then, another O to 2.5 µl of distilled water, depending on the body size, was added 
to each well. Up to 13 runs, yielding 16 loci, could be obtained from each 
individual. However, the PEP-2 locus stain was too light to be read accurately, 
so this locus is not included in the analyses. 
Buffer systems follow Richardson et al. (1986). Fifteen loci were resolved 
by an injtial survey of 27 enzyme stains across 11 buffers. The optimal buffers and 
running conditions for each locus, used in data collection, are given in Table 1-4. 
Gels were soaked prior to loading in the same buffer as for running. They were 
run in a refrigerator at 180 volts for 35 to 90 minutes (Table 1-4 ). 
Staining was carried out using an agar overlay, following recipes taken from 
Hebert and Beaton (1989). The agar was washed away after the gel was 
sufficiently stained. Gels were then soaked in water for at least half an hour and 
recorded by drawings, photographs or xeroxing. All gels were preserved dry in 
23 
transparent pocket sheets and filed for later reference. 
Fourteen loci were recorded for the adults. These plus PEP-1, discovered 
later, were analyzed for nymphs. Three of the loci were monomorphic m 
Pachypsylla, but the other 12 showed differences among gall types. 
Electrophoretic data analysis: 
With exceptions to be noted, alleles at each locus were originally 
designated alphabetically from fast to slow in the adult data and (separately) in 
the nymphal data for leaf gall makers. The codings for adults and nymphs do not 
correspond exactly, because the adult were analyzed first, and additional rare 
alleles were found in nymphs. Moreover, some of the alleles originally recognized 
at five loci (PGM, ME, PEP-1, MDH-1 and LDH) in some gall makers were 
subsequently found to consist of two distinct alleles. The new alleles were given 
new codes, out of alphabetical order (see footnote to Table 1-6). 
The fit to Hardy-Weinberg equilibrium was analyzed at each variable locus 
using the Biosys-1 package of Swofford and Selander (1981). Chi-square 
contingency table analyses, carried out in Systat 5.2, were used for testing the 
heterogeneity of allele frequencies for all loci among samples. Alleles for which 
the expected value was less than five in more than 1/5 of the cells were pooled 
(Sokal and Rohlf, 1981). Multiple pairwise, a posteriori, comparisons were carried 
out, using the adjusted a' = 1- (l-a)1/k (Sokal and Rholf, 1981, p. 728), where k 
is the number of comparisons. This value was rounded down to the nearest value 
24 
.--rar-r~,._,,.,,-- -
table in Rohlf and Sokal (1969), making the test more conservative. 
Life history traits: 
The life histories of two leaf gall makers, the glabrous nipple gall and the 
star gall, in the NAL population were observed closely starting in 1990. I visited 
the NAL tree every two to four weeks from March to November, 1991, in addition 
to occassional observations made at various localities between 1990 and 1994. 
Collected galls were dissected and preserved in alcohol or by freezing. The 
developmental stages of the insects were recorded for each gall type. 
Rearing experiments: 
Individual wild-caught females were caged on hackberry seedlings in the lab 
to determine whether their progenies are always homogeneous in gall type. The 
sources of insects were two nearby populations in Maryland in which the same two 
gall types occur. In one, the NAL population, the glabrous nipple gall (LNg) is 
more abundant than the star gall (LS), whereas in the other, the BB population, 
the reverse is true. No absolute morphological differences are known between the 
adults of LNg and LS, but the adults emerging from the two gall types differ in 
size. To ensure that both gall types were represented in the experiment, adults 
Were sorted into "large" (bigger than 3mm), presumably LNg, and "small" (smaller 
than 3mm), presumably LS, size classes. Only dark abdomen females were used, 
to avoid side cell inquilines (see Chapter II). Species distinctness is supported if 
25 
-:""r l 
I 
/; 
1/' 
I 
/, 
I 
I 
siblings always have the same gall type, but progeny from different individuals set 
up on the same date are different in gall type. 
Cylindrical cages (30 cm X 50 cm) were sewn from white polyester fine-
mesh organza, incorporating a median belt of transparent mylar ( 4cm wide, 0.012 
mm thick) to permit convenient inspection of the cage contents. These cages were 
slipped onto branches and tied on with "Stretchrite" round cord elastic at both 
ends. 
Originally, this experiment was intended to be carried out in the field. 
Numerous cages were set in place in the field before budbreak to exclude natural 
infestation by adult psyllids. Single females were introduced into the experimental 
cages when the insects became available, while none were introduced into control 
cages. However, several weeks into the experiment it was apparent from the 
numbers of galls in both control and experimental cages that the cages had failed 
to exclude natural infestation. The field experiment was therefore abandoned, and 
a reduced version of the original design was added to a lab rearing experiment 
conducted to investigate the possibility of an inquiline species (see Chapter II). 
In the lab experiment, wild-caught females were caged on potted seedlings, 
three or more years old, during the period when both types of galls are being 
initiated in the field. The seedlings included both Celtis tenuifolia and C. 
occidentalis and also varied in leaf age. The latter had been manipulated to test 
the possibility that differences in gall morphology are due not to differences in the 
cecidogen, but to attack of the host during different stages in leaf development. 
26 
-
"Old" (well developed) versus "young" (still-developing) foliage seedlings were 
produced by covering the plants with plastic bags, placing wet paper towels at the 
bottom of the bag, keeping the plants in a cold room (2-5°C) starting in late 
winter, and moving them back to the lab at different times the following spring 
(1993). 
Seedlings were used as their foliage reached the appropriate stage of 
development. Four sets of cagings were set up, two using females from the NAL 
population (May 13 and 24 ), both on C. tenuifolia, and the other two with females 
from the BB population (May 17 and 23), one on C. tenuifolia and the other on 
C. occidentalis (see Table 1-9). Four or five seedlings were used each time, for 
a total of 18. Each seedling received three cages, one containing a large female, 
another a small female, and the other had no psyllids as a control. (Each seedling 
also bore three other cages, part of an experiment on inquilines - see Chapter II.) 
The seedlings were kept near a south-facing window in ambient light. Females 
were removed and frozen at -80° C ten days after being caged. The type of galls 
formed within each cage was recorded one month after infestation. 
Morphological methods: 
Morphological comparisons were conducted using both optical and scanning 
electron microscopy. Color variation of live insects was recorded immediately 
after dissection from the gall. Specimens were preserved in a freezer, in alcohol 
or dry. Specimens were dissected and slide mounted in Canada balsam when 
27 
necessary. Both adults and last (5th) instar nymphs were examined. Taxa 
examined included gall makers from all five type of leaf galls, and petiole, bud 
and twig galls. Ten to thirty specimens were examined for each gall type and each 
stage, except that for rare species only 3-7 individuals were used. The specimens 
inspected for each taxon included 2-7 populations across geographic regions 
(Table 1-3). 
Body sizes of gall makers from petiole, glabrous bud and all five leaf gall 
types were measured based on electrophoretic samples. The sample size of each 
gall types were 42 to 99 for leaf gall makers and 14 to 15 for non-leaf gall makers. 
The body size variation among leaf gall makers was analyzed using Tukey-Kramer 
methods of multiple comparisons among pairs of means (Sokal and Rohlf, 1981, 
p. 252). 
Specimens for scanning electronic microscopy were sonicated before 
dehydration. Five treatments of specimen preparation process were compared 
using 2 individuals each in a preliminary study. In the full process, specimens 
were pre-treated with glutaradelhyde fixer (4%, 12hrs) and osmium tetroxide (2%, 
4hrs), dehydrated in a series of increasing ethanol concentrations (30, 40, 50, 60, 
70, 80, 90, 95% each 10 mins and 100% EtOH 1hr), then critical point dried in 
CO2 using a "Samdri-780A" critical point dryer. Three other treatments were 
similar but the pre-treatment of glutaradelhyde and osmium tetroxide was 
eliminated. Each of these treatments started with different concentrations of 
ethanol (30-40-50-70-80-90-100%; 70-80-90-95-100% and 70-100%), followed by 
28 
-' 
critical point drying. The final treatment was air drying alone. Air-dried 
specimens shrank dramatically and this method was abandoned. No significant 
differences were found between specimens which were or were not pre-treated 
with glutaradelhyde and osmium tetroxide. Neither did it matter whether 
dehydration started with a lower EtOH concentration (30%-) or a higher (70%-). 
The final procedure adopted was an EtOH series with concentrations of 70-80-90-
95-100%, followed by critical point drying. Specimens were then mounted on 
aluminum stubs with double-sided adhesive tape, sputter coated with gold-
palladium using Hummer V and examined with an Amray scanning electronic 
microscope. Polaroid 4X5 photos (type 55, black and white) were taken. Two to 
four adults and nymphs of each gall type were examined. 
RESULTS 
1. Do galls on different plant parts represent different species? 
Electrophoretic analyses of the 1991 adult (Table 1-5) and 1992 nymphal 
(Table 1-6) samples confirm that psyllids that make petiole, bud and twig galls are 
clearly separate species, from each other and from leaf gallers. There are 
multiple fixed or nearly fixed allelic differences among these classes of gall types. 
For example, allele C (adult) or D (nymph) of TPI is fixed in and unique to the 
petiole gall maker, allele D (adult) or E (nymph) of TPI similarly distinguishes the 
bud gall makers, and allele D of GPDH and allele C of TPI characterize the twig 
gall maker (Tables 1-5 and 1-6). 
Variations among populations within the glabrous bud galls were minor, 
29 
suggesting that it represent a single species (Table 1-7, contrast AOO). When the 
hairy bud gall was added, the differences were significant (Table 1-7, contrast AO). 
However, five out of the six pairwise X2 contingency tests of allele frequency 
between the four bud gall populations in the nymphal data showed no significant 
differences (Table 1-7, contrasts Al-A6). The only significant pair was between 
glabrous bud gall from Virginia and hairy bud gall from Oklahoma. Since the two 
populations were far apart from each other, the geographic variation may 
contribute to the variation. However, the two types of bud galls, hairy versus 
glabrous, have some times been regarded as distinct (Crawford, 1914; Tuthill, 
1943). Tree building methods usually intermingle these two bud gall types (see 
Chapter III, Fig. 3-2). 
The three geographic populations of petiole gall makers show significant 
differences in allele frequencies when considered all together (Table 1-7, contrast 
BO) or pairwisely (Table 1-7, contrasts Bl-B3). No morphological differences 
were found among these populations except that the body color of the last instar 
nymphs from Arizona is lighter (yellow) than the color (orange) of those from the 
other two populations. 
Gall makers attacking different plant parts are also distinct in morphology. 
Although the bud and twig gallers are very similar, one can easily distinguish them 
by body size and wing maculation. 
The petiole gall maker has a large body size including wings ( ca. 5 .5-6.5 
mm) compared to the rest of Pachypsylla, which range from 1.9 to 4.4 mm (Table 
30 
1-8). The forewing of the petiole galler is clear with a few brown maculations, 
mostly in the apical region (Fig. 1-5). The glabrous and hairy bud gallers are very 
similar. Both are of moderate size ( ca. 3.2 mm). However, while the forewings 
of the glabrous gall type are uniformly brownish, the hairy type has a lighter 
brown background with an oblique clear band subapically. The twig galler looks 
like the bud galler but its body size is bigger (3.5-4 mm) and the brown colored 
region is restricted to the margin of the distal half of the wing. The leaf gallers 
range in body size from 1.9 to 4.4 mm. Their wings are uniformly mottled and 
spotted except for a clear subapical oblique band. 
The bud and twig gall makers also have similar life histories. Their adults 
emerge in late spring and overwinter in the last (5th) nymphal instar. The petiole 
gall makers also overwinter in the last (5th) nymphal instar but the adults emerge 
in very early spring. The leaf gall adults emerge in the fall and it is this stage that 
overwinters. 
2. Do the different leaf gall morphs represent different species? 
Allozyme evidence 
Initial evidence on this question came from the electrophoretic analysis of 
adults collected in 1991 (Table 1-5). These data suggest that species limits will 
be harder to determine within the leaf gallers than between these and other gall 
positions. The Nei genetic distances among leaf gallers are all smaller than 0.1, 
and there are no fixed or nearly fixed differences (Table 1-5; see also Fig. 3-3 a). 
31 
However, there are almost always significant frequency differences between 
different gall types from a single locality. For example, at Great Falls, Virginia, 
there are five co-occurring gall types on Celtis occidentalis. Three of these types 
were collected from the same individual tree. At this site, all 10 pairwise 
comparisons between leaf gall types showed strongly significant differences (Table 
1-9, contrasts Cl-ClO). This suggests that the leaf gall makers do not form a 
single randomly mating population. 
The hypothesis that leaf gall morphs represent separate species would be 
stronger if one could show consistent frequency differences of particular alleles 
' 
in comparisons from different locations. The most consistent differences are those 
between the blister gall maker and other gall makers. For example, adults show 
a consistently much higher frequency of the C allele and a lower frequency of 
allele E in IDH-1 in the blister galler than in other sympatric gall types, in 
collections at three widely separated localities from the same host, C. occidentalis 
(Table 1-10). A similar trend was seen at the 6PGDH locus. 
A difficulty with interpreting the adult data is that in the adult samples, 
center-, mono- and side-cell individuals are combined. The side cell nymphs 
actually represent a distinct species (see Chapter II). Their occurrence could 
either decrease the ability to detect frequency differences between gall types, or 
produce apparent differences where none exist, depending on the relative 
frequency of side cells among different gall types. However, removal of adult 
individuals carrying the allele E ( equivalent to allele F in the nymphs) at malic 
32 
" ·""' 
I I 
enzyme, the strongest marker for the side cell nymphs (see Chapter II), had no 
effect on the conclusions to be drawn from Table 1-10 (contrast set C), suggesting 
that the differences among gall types are not due to "contamination" by the side 
cell form. 
Direct support for the reality of the patterns seen in the adult data comes 
from analysis of nymphal collections in 1992, from which inquilines were excluded 
by using monocell or center cell individuals only. As in adults, the allele 
frequencies in sympatric gall morphs are generally significantly different. For 
example, the allele frequencies of the three leaf gall types ( disc, blister and 
glabrous nipple) types sampled from Great Falls, Virginia were significantly 
different from each other pairwisely (Table 1-9, contrast Dl-3), paralleling the 
adult findings. Similar differences were detected between the two gall types 
sampled from Catoctin Mountain (Table 1-9, contrast El), and the other two gall 
types found in Maryland (BB and NAL) populations (Table 1-9, contrast Fl). 
Concordant with the adult data, the allele frequencies of the blister gallers 
differed most obviously from other gallers, consistently across localities, in the 
frequencies of allele Din IDH-1. A similar pattern, though less pronounced and 
consistent, is found in 6PGDH, as in the adult data. 
Life history evidence 
Phenological differences were found among co-occurred leaf gall makers 
in the National Agricultural Library population. Of the two types of leaf galls 
33 
which co-exist on this tree, even on the same leaf, nipple galls were formed earlier 
than star galls (Fig. 1-5). Eggs of glabrous nipple galls were laid earlier in the 
spring than those of star galls. 
Rearing evidence 
In 18 of 36 cages set up with single females, three or more leaf galls were 
found (Table 1-11). None were found in the other 18 cages, presumably because 
the female either died or was unfertilized at capture. None of the control bags 
had galls at all. In every cage containing galls, all the galls, necessarily 
representing progeny of a single female, were of the same type. Seven cages 
contained nipple galls while eleven had star galls. In 14 of 18 cases, the gall type 
was that predicted from the size class of the female. Both types of gall were 
formed on both "old" and "young" foliage class seedlings, and on two seedlings, 
both types of galls occurred (in separate cages). This suggests that foliage age is 
not a determinant of gall shape. Progenies of "large" females did not always 
produce nipple galls, nor did those of "small" size females always produce star 
galls, suggesting either that 3 mm is not the appropriate recognition criterion, or 
that the sizes of these gall-formers overlap. However, there is a strong association 
between the body size and type of galls made (see foot note to Table 1-11 and 
next section). 
34 
ii: 
I/ 
I 
' 
Morphological evidence 
Morphologically, the leaf gallers are extremely similar to each other, with 
highly overlapping variation. All the differences I could find were relative, not 
absolute. However, measurements of adult body size among almost all different 
pair leaf gall types show significant differences, except those of two types of nipple 
galls. The galler which is most nearly distinguishable from the others is the blister 
gall maker. The smaller body size ( < 2.5mm) of the blister gall maker was used 
to separate it from the nipple galler (3-4rnm) by Crawford (1914) and Tuthill 
(1943). The male genitalia of the blister galler are also slightly different from 
those of the others: the male proctiger is less hairy and the forceps are more 
sharply pointed (Fig. 1-7). 
All other gallers are very hard to identify without knowing the gall shape. 
3. Do Pachypsylla of similar gall type on different host species represent different 
species? 
Allozyme allele frequencies did not differ between blister gall nymphal 
samples from different localities on the same host species in Texas. The data 
were therefore pooled to yield a single contrast between the two host species 
(Table 1-6, populations 10 & 11). There were no significant frequency differences 
between these populations, either at individual loci or when they were combined 
(Table 1-9, contrast Gl). Thus, there is no evidence of allozyme divergence 
between sympatric blister gall makers from the two different host species. 
35 
Morphological differences between blister gall makers from C. laevigata, C. 
reticulata and C. occidentalis were also investigated, especially in the male 
aedeagus, following Riemann (1961). Riemann found the largest differences to 
lie in the head/shaft ratio of the terminal segment of the aedeagus. He reported 
ratios of 0.5, 0.25 and 0.38 for blister gallers in Texas and Oklahoma from C. 
laevigata, C. reticulata, and C. occidentalis, respectively. My measurements on 
blister galler specimens yielded mean estimates of 0.50, 0.33, and 0.33 for the 
same ratios. The former two were from Texas and the latter was from Maryland 
and Virginia. These means are similar to Riemann's except for specimens from 
C. reticulata. The differences between blister gallers from C. laevigata and C. 
reticulata in my samples are significantly different from each other (see foot note 
of Table 1-12). I also measured blister gallers from Arizona, where only Celtis 
reticulata is available. The mean ratio and range are similar to the blister gallers 
from Texas, though the sample size of Arizona population is small (Table 1-12). 
No other significant morphological differences between blister gallers from 
different host plants were found. 
DISCUSSION AND CONCLUSIONS 
Evidence from morphology, allozymes and life histories strongly confirms 
that the petiole, bud, twig and leaf gall makers each constitute at least one species 
distinct from the others. 
Within the bud gallers, allozymes show little geographic variation, even 
36 
between allopatric gall morphs, which argues against the separate species names, 
P. celtidisgemma Riley (glabrous bud gall) and P. pallida Patch (hairy bud gall) 
advanced for these morphs by some authors. However, the gall morphs are 
associated with differences in wing pattern and host affiliation. Within the petiole 
galls, there is significant allozyme differentiation among geographic populations 
on different hosts. The possibility that allopatric populations of the petiole and 
bud gallers are specifically distinct remains open. 
The members of the leaf gall making complex are very closely related and 
hard to distinguish. However, several lines of evidence indicate that they are at 
least partially reproductively isolated, and quite possibly separate species. First, 
sympatric gall types always show statistically significant differences in allozyme 
frequencies. In the case of the blister gall, the differences are marked enough, 
and consistent enough across geography, to argue by themselves for strong 
reproductive isolation, supporting the hypotheses of Crawford (1914) and Tuthill 
(1943). Second, the rearing experiment, in demonstrating that gall shape variation 
between isofemale progenies is both all-or-none and associated with body size, 
provides evidence at least strongly consistent with a genetic, species-specific basis 
for that trait. Third, pairs of sympatric gall morphs examined in detail differ 
substantially in the timing of adult appearance and/ or gall formation as well as 
body size. Though more evidence is needed, it is probable that in addition to the 
blister ga.ller, the thTee species named by Riley but yncmymized by others (e.g. 
Crawford, 1914; Tuthill, 1943) , i.e. P. ,c,eUidisumb.il.icus (disc g er , P. 
37 
celtidisglobulus (glabrous nipple galler) and P. celtidisasterisca (star galler), should 
be revived. 
Allozyme variation between sympatric blister galls from C. laevigata and C. 
reticulata is small, which casts doubt on Riemann's hypothesis that blister gallers 
on different hosts represent different species. In contrast, significant 
morphometric differences in the male genitalia, corroborating Riemann's 
observation, seem to support that hypothesis. Since morphometric variation is 
subject to environmental effects including host differences, however, it is not clear 
whether the observed differences have a genetic basis. Reciprocal host transplant 
experiments are needed to answer this question, and the species issue remains in 
doubt. 
Conclusions about the status of several additional described nipple gall 
makers, including P. celtidispubescens, P. rohweri, P. celtidisglobulus, P. 
celtidiscucurbita and P. celtidiscucurbita var."?", can not be made, since they were 
not included in this study. 
Further studies using other techniques and more extensive and intensive 
geographical sampling are needed to fully resolve the species problem among leaf 
gall makers. One constraint of the cellulose acetate gel electrophoresis used here 
is that the small size of the gel does not allow enough migration distance to 
separate alleles of very similar mobility. 
38 
KEY TO SPECIES OF PACHYPSYLLA STUDIED 
1 Vertex and thoracic dorsum not shining, covered with conspicuous, short 
stiff pubescence ....................................... 2 
Vertex and thoracic dorsum glabrous, with only sparse, minute 
pubescence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 
2(1) Body size large (5-6 mm); forewing narrowly rounded at the apex, clear 
with maculation mainly apically; galls spherical and large (1-3 cm) on 
petiole, usually polythalamous . . . . . . . . . . . . . . venusta Osten-Saken 
(petiole gall maker) 
Body size small to moderate (2-4 mm); forewing broadly rounded at the 
apex, mottled and spotted throughout, with a clear subapically oblique 
band; galls on leaf-blade of various shapes, monothalamous or 
polythalamous ........................................ 3 
3(2) Female abdomen; male forcep tip roof-like, side margins parallel, straight 
in lateral view; nymphs found in side cells of other leaf-blade galls; 5th 
instar nymphs with yellow wing pad . . . . . . . . . . . . . . . . . . . . . . . . . . 
............................ cohabita sp.n Yang and Riemann 
(leaf inquiline psyllid) 
Female abdomen brown; male forceps in lateral view gradually tapering 
in apical half, tip pointed, side margins convex; nymphs found in single 
cell galls or in center cell of multiple-cell galls, 5th instar nymphs with 
brown wing pad . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 
39 
4(3) Gall on the upper side of leaf, raised or convex .............. . 5 
Gall on the upper side of leaf concave, forming a depression . . . . . 6 
5( 4) Galls inconspicuous, blister like, elevated on both upper and under sides 
of the leaf . .. ........................... celtidisvesicula Riley 
(leaf blister gall maker) 
Galls rounded, raised on the upper side; star-shaped or flower shaped on 
the under side . . . . . . . . . . . . . . . . . . . . . . . . . . celtidisasterisca Riley 
(leaf star gall maker) 
6( 4) Galls disc-like or button-like, much flattened; upper depression with an 
outer rim and the central portion slightly raised with a median spine; 
under side wart-like, with a depression at middle . . . . . . . . . . . . . . . . 
celtidisumbilicus Riley 
(leaf disc gall maker) 
Galls nipple-like, much rounded; upper side with depression and under 
side rounded . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 
7(6) Galls glabrous; upper side forming a cup-like depression with an outer 
rim; under side rounded, nearly as long as wide or wider than length, 
apple-like, with a central depression . . . . . . . . . celtidisglobulus Riley 
(leaf glabrous nipple gall maker) 
Galls pubescent on the apical portion of the under side, but hairs easily 
rubbed off, giving glabrous appearance; upper side a depression with 
central spine but no outer rim; under side nipple-like, longer than wide, 
40 
without obvious central depression .......... celtidismamma Riley 
(leaf hairy nipple gall maker) 
8(1) Pterostigma prominent; wings with a smoky band along the anal and 
apical margins and extending along veins toward base; galls oblong-oval 
in shape on one side of the twig, monothalamous 
celtidisinteneris Mally 
( twig gall maker) 
Pterostigma not obvious; forewing homogeneous or maculated not as 
above; forming irregularly round galls on bud ................. 9 
9(8) Forewing uniformly brown; gall glabrous on bud, usually polythalamous 
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . celtidisgernma Riley 
(glabrous bud gall maker) 
Forewing with light brown in apical half extending to some basal regions; 
gall densely pubescent on bud, usually polythalamous . . . . . . . . . . . . 
........................................... pallid a Patch 
(hairy bud gall maker) 
41 
Chapter II 
The evolution of inquilinism in hackberry leaf galling psyllids: 
Are multiple individuals within the same gall conspecific? 
42 
INTRODUCTION 
The organisms and interactions that occur in a gall are often more diverse 
than one might expect. A gall very often represents a complex community rather 
than a microhabitat for a single gall former. Some oak galls, for example, support 
as many as 22 species of insects (Askew, 1961), comprising the original gall 
former, inquilines and their parasites. Contemporary studies have shifted 
emphasis from rearing and identifying the emergent insects of a gall to 
understanding the ecological relationships of the gall community (Wiebes-Rijks 
and Shorthouse, 1992) and their evolutionary origin (e.g. Ronquist, 1994 ). 
However, most such works have focused on major galling groups, e.g. Cynipidae 
(e.g. Askew, 1961; Brookfield, 1972; Shorthouse, 1973; Washburn and Cornell, 
1979; Abe, 1992) and Cecidomyiidae (e.g. Askew and Ruse, 1974; Roskam, 1979, 
1986; Roskam and Van Uffelen, 1981; Shorthouse and West, 1986). 
Only a few studies have examined the relationships of inhabitants of psyllid 
galls (e.g. Jensen, 1957; Walton, 1960; Moser, 1965; Hodkinson and Flint, 1971) 
and all concentrated on the natural enemies of the gall formers. No inquilines 
have been reported in psyllids except Riemann's (1961) observation on hackberry 
psyllids, Pachypsylla, in his unpublished dissertation. Present research tests 
Riemann's and related hypotheses on the potential inquilinism in Pachypsylla, and 
discusses the evolution of this trait. 
43 
BACKGROUND 
Gall Inquilines 
The term "inquiline" comes from the Latin word inquilinus meaning 
temporary inhabitant or guest (Brown, 1956, pp. 439 & 387) and has been broadly 
used. For instance, a search of the Life Science Collection on CD-Rom 
(Cambridge Scientific Abstracts, Bethesda, MD) found 57 papers on "inquilines" 
between 1982 and 1993. The "inquiline" in these research projects ranged from 
va rious kinds of organisms in galls, to social parasites in Hymenoptera (e.g. 
Bourke and Franks, 1991; Reed and Akre, 1983), to crickets (e.g. Henderson ancl 
Akre, 1986; Bolton, 1986), beetles (e.g. Howden and Gill, 1988) ancl worm lizards 
(e.g. Riley et al., 1986) in ant nests, and to co-occurring fishes in coral reefs 
(Spiegel, 1980). For the purposes of this study, the term will be restricted to gall 
inquilines unless otherwise indicated. 
While reports of inquilinism are rare in psyllicls, inquilines are frequent in 
other galling groups. Gall inquilines are gall inhabitants that feed on ga ll ti ssues, 
without directly damaging the gall-inducer, but are unable to induce gall s 
themselves independently (Skuhrava et al., 1984; Shorthouse and West, 1986; 
Shorthouse, 1991; Wiebes-Rijks and Shorthouse, 1992). Although inquilines can 
be viewed as parasites of gall tissue, their ecological relationship to the original 
gall former is not clear, and should not be assumed to be parasitic. Meyer ( 1987, 
p.177) defined inquilines as "insects living commensally in the ga ll cortex," 
implying that they do little harm to the original gail inducer. The inquiline 
44 
Synergus pallicomis in galls of the cynipid Cynips quercusfolii is an example (Askew, 
1961). However, other studies on cynipid galls have demonstrated that inquilines 
have negative effects on gall inducers by food deprivation, or even by killing the 
host at an early stage of development of the gall (Evans, 1965, 1967; Shorthouse, 
1973, 1980; Washburn and Cornell, 1981). For example, the inquiline Periclistus 
kills the gall former Diplolepis at oviposition (Shorthouse, 1980). 
In other homopterans, gall inquilines have been reported in the aphids 
genus Eriosoma (Akimoto, 1981, 1988). Akimoto hypothesized that facultative 
inquilinism is a pre-adaptation for obligate inquilinism. He postulated that the 
latter arose by dispersal beyond the range of the primary host. This could result 
in strong selection for invasion of galls formed by related species, if the 
dispersants could not form their own galls on the available hosts. Subsequent 
specialization for this habit might preclude the re-acquisition of gall induction, 
even on re-encounter with the ancestral host. 
Occurrence of multiple individuals in Pachypsylla leaf galls 
The number of nymphs within a single hackberry psyllid gall varies from 
one to 17 and probably more, depending on the gall type and population. Petiole 
and bud galls are usually polythalamous, that is, with multiple individuals in the 
same gall, and each is confined to a separate chamber. Twig galls are usually 
monothalamous, that is, with only one individual per gall. 
The situation in leaf gall makers is more complicated. Usually there is only 
45 
one individual inside a gall, but sometimes, particularly in the nipple gall in 
Maryland and vicinity, there can be multiple individuals, each enclosed in a 
separate cell. The proportion of multiple-cell galls and the number of cells within 
one gall vary between gall types and localities. 
There are no obvious differences among individuals and cells within the 
same petiole or bud gall. In contrast, within multiple-cell leaf galls one can 
generally distinguish between a center cell, presumably the individual that initiated 
the gall, and one or more "side cells." This distinction raises a series of questions. 
Are multiple individuals within the same gall conspecific? Does the 
differentiation between the center and side cells within the same leaf gall suggest 
that there is more than one species? Do they both contribute to gall production? 
Do the side cell and center cell individuals differ in performance and 
survivorship? What are the ecological relationships between the center and side 
individuals? What is the adaptive significance of having monothalamous or 
Polythalamous galls? If the side cell individual is a true inquiline, how did it 
evolve? This study was designed to answer some of these questions. 
Two interpretations of the species status of side cell individuals within 
Pachypsylla leaf galls have been offered. Moser (1965) observed multiple cells in 
leaf hairy nipple galls on Ce/tis occidentalis in New York State. He called the side 
cell a "marginal gall" and considered the side cell nymph to be conspecific with the 
blister gall nymphs, the only other leaf gall that occurred in that area. 
Riemann (l96l) found the side cell individuals in several different leaf gall 
46 
types in Texas and considered the side cells species to be undescribed. He 
regarded the side cell nymphs, across different gall types, to be the same species 
but not conspecific with the various center cell species. He hypothesized that the 
side cell species was unable to induce a gall and is an inquiline that became 
incorporated into the gall by feeding next to the gall maker during gall initiation 
in the spring. He also suggested that the inquiline sometimes killed the gall 
maker by expanding its own cell too much. 
In my initial observations, there were consistent color differences between 
the fifth instar nymphs from center cells and side cells, especially in wing pad 
color. The center cell nymph always has darker wing pads, which are usually 
brown, whereas the side cell nymph always has light colored wing pads, which 
usually are yellow. The nymphs from mono-cell galls and from center cells within 
multiple cell galls are either brown in general color with green or red color in the 
intersegmental membranes while the side cell nymphs are greenish yellow in 
general body color without differentially colored intersegmental membrane. 
Field observations and lab rearing results also suggests that first-instar 
nymph and adult coloration are associated with cell position, at least in the 
glabrous nipple gall and the star gall in the Maryland area. Side cell nymphs in 
the first instar are white, sometimes with dark maculation, and adult females have 
a green abdomen. Center cell individuals have yellow first instar nymphs and the 
adult female has a dark abdomen. 
These observations indicate that there is considerable variation between 
47 
side and center individuals besides the position in the gall and suggest that side 
cell nymphs may be a different species than center cell nymphs. 
QUESTIONS ADDRESSED 
Two questions were asked in this study: 
1. What is the nature of side cell individuals within a leaf gall? 
The following hypotheses are contrasted: 
A. The null hypothesis is that the side and center cell individuals are 
the same species. 
B. Moser's hypothesis predicts that in multiple cell nipple galls, side 
cells are "marginal galls" that contain individuals of the co-occurring 
blister gall maker that have become incorporated into the nipple 
galls. 
C. Riemann's hypothesis predicts that side cell individuals represent a 
separate species, that does not have the ability to form its own gall. 
2. Are side cell individuals from different types of leaf galls conspecific? 
Riemann's hypothesis is that there is only one side cell species regardless 
of the "host" species. There are two other possible hypotheses 1) that each 
kind of leaf gall has a separate inquiline species or 2) that the inquiline is 
the same species as the host. 
Samples of side cell individuals from two types of nipple galls, glabrous and 
48 
hairy, were examined morphologically and electrophoretically to test these 
hypotheses. A rearing experiment was designed to test the ability of side cell 
individuals to induce galling. Dependency of the side cell individuals on the 
center individual was further investigated by destroying the center individual at 
different stages of gall initiation. 
MATERIALS AND METHODS 
Samples for electrophoretic and morphological studies 
Center versus side-cell nymphs from the two types of nipple galls in 
Maryland were compared to each other and to those from single-celled galls of 
other co-occurring leaf gall types using allozyme, morphological, and life history 
characters to assess possible species differences. 
Nymphs were examined from the hairy nipple gall and glabrous nipple gall 
from three populations in Maryland in 1992, plus two other types of galls that 
commonly co-occur with these nipple galls. These included hairy nipple galls and 
blister galls from Catoctin Mountain (CMt) in northern Maryland (population #1-
4 in Table 2-1), and glabrous nipple galls and star galls from both the National 
Agricultural Library (NAL) population in Beltsville (population #5-8 in Table 2-1) 
and the Branchville Road (BB) population in Berwyn (population #9-12 in Table 
2-1). In both locations, the host is Ce/tis tenuifolia. The NAL population occupies 
a tree about 12 meters high located beside a small graveyard in the middle of an 
open lawn. Few seedlings were found near it but there is another tree 3 m high, 
49 
500 m from it. The BB population contains many hackberries, from small 
seedlings to tall trees, in a residential neighborhood. The two populations are 
about 5 km apart. The CMt population is located at the Poplar Grove campsite 
in Catoctin Mountain Park where trees and seedlings of C. occidentalis are 
abundant. This population is at about 450 m elevation and about 85 km from the 
BB and NAL populations. 
The sympatric non-nipple gall is the blister gall, which never has multiple 
cells and often co-occurs with the hairy nipple gall, and the star gall, which often 
co-occurs with the glabrous nipple gall. Galls were collected in the fall of 1992 
and 1993 from at least two trees in BB and CMt and one tree in NAL. I dissected 
each gall, separated the side cell nymphs from the center cell nymph, and 
collected the nymphs from mono-cell galls. These nymphs were either frozen for 
later electrophoretic and morphological studies, or reared in small glass vials to 
obtain the associated adults for morphological comparisons. 
Electrophoretic and morphological methods 
Three to twenty-one nymphs from each cell position in each locality were 
analyzed by electrophoresis (Table 2-1 ). Five to twenty individuals of each 
category were examined for the morphological comparison. Electrophoretic, 
morphological, and data analytical methods are as described in Chapter I. 
Allozyme frequency data for samples from the same gall type and cell position 
were combined if they did not differ significantly. 
50 
Rearing experiments 
Two experiments were conducted in the spring of 1993 to test the inquiline 
hypotheses. The first experiment tests whether the side cell nymphs can form 
galls, and also provides further data on color differences between center and side 
cell individuals. Fertilized females were collected from the NAL and BB 
populations and confined in cages on hackberry seedlings in the lab. This 
experiment was carried out at the same time and on the same plants as described 
in Chapter I. 
There were three treatments: 1) cages containing a single brown abdomen 
female, presumably a center cell individual (these are the same females described 
in the rearing experiment in Chapter I); 2) cages containing a single green 
abdomen female, presumably a side cell individual; and 3) cages containing one 
brown plus one green abdomen female. In treatment 3, both "large" ( > 3mm) and 
"small" ( <3mm) brown abdomen females were used (in separate replicates), to 
ensure inclusion of both star and nipple galls, as described in Chapter I. Control 
bags contained no insects. 
Each replicate was reared in a separate bag, but at least one replicate of 
all treatments and one control were reared simultaneously on the same seedlings. 
Thus, each seedling had at least six cages, each contain: a single large brown 
abdomen female, a single small brown abdomen female, a single green abdomen 
female, a pair consisting of one large brown abdomen plus one green abdomen 
female, one small brown abdomen plus one green abdomen female, or no psyllid 
51 
at all. The number and timing of the replicates set up, shown in Table 2-5, 
corresponds to that previously described (Table 1-9). Most seedlings used were 
C. tenuif olia, except that one batch of insects was reared on C. occidentalis 
seedlings. 
Females were removed from the cages after ten days and frozen at -80°C, 
and the color of nymphs found was recorded. Gall formation or lack thereof was 
recorded a month after infestation. Galls were dissected in September to examine 
cell numbers within each gall. 
Riemann's hypotheses predicts that treatment 1 will have yellow nymphs 
and mono cell galls only, treatment 2 will have white nymphs and no galls, and 
1 ·1, 
I I 
treatment 3 will have white and yellow nymphs, and both mono cell and multiple I/! 
I !! 
I 
cell galls. 
Destruction experiment 
The second experiment tests the dependency of side cell on center cell 
nymphs. Aggregates of newly hatched nymphs were located in the field and the 
center nymph experimentally destroyed. Destruction was conducted under a 
dissecting microscope fixed on a plastic manipulation and recording platform, 
which could be either fastened on a tripod or suspended from the shoulders on 
straps, leaving hands free for dissection and recording. Insect pins were used to 
pick the center cell nymph out of the leaf. 
52 
Treatments were: 1) center cell nymphs destroyed early, i.e., before any 
nymphs were enclosed in the gall; 2) center cell nymphs destroyed after the center 
cell individual was enclosed in the gall but the side cell nymphs were still exposed; 
3) center cell nymphs destroyed when both the center and side cell nymphs were 
fully enclosed; 4) control, in which galls in developmental stages corresponding to 
the first three treatments were left undisturbed. Predictions were that treatment 
4 would develop normally, galls of treatment 1 would not develop and side cell 
nymphs would die, while galls in treatment 2 and 3 might continue their 
development and side cell nymphs might survive. 
Experiments were initiated during four time intervals in late May, in the 
NAL population. Leaf gallers were abundant, and several nymphal aggregations 
in different gall development stages could be found even within a single leaf. 
Each replicate, consisting of two or more treatments, was performed on a single 
leaf when possible, or otherwise on different leaves on the same twig. The 
experimental leaves were numbered and marked with blue plastic tape on the twig 
right below where the petiole attached. The position of each gall on the leaf was 
sketched on paper when destruction was carried out, and the development (size) 
of the gall was measured immediately and again two weeks, two months, and four 
months after destruction. The total number of replications for each treatment, 
which was determined in part by the difficulty of finding galls in all fou r 
developmental stages, ranged from 16 to 30. 
53 
Female phenology 
To compare the phenology of center versus side cell females, females were 
collected in the NAL population every five days in spring of 1993, from April 15 
to May 15. Females found on the plant were randomly collected from different 
branches. About 50 to 70 females were collected each time, and the numbers of 
brown abdomen and green abdomen individuals recorded, to determine the 
change in proportion of the two morphs over time. 
RESULTS 
Electrophoretic data 
Allozyme frequencies in center cell nymphs from multiple cell galls and 
nymphs from mono-cell galls of the same gall type in the same population were 
not significantly different (Table 2-2, contrast Al-A3), so these data were pooled. 
I also pooled the two populations (NAL and BB) of combined mono and center 
cell nymphs of glabrous nipple gall makers (Table 2-2, contrast B 1 ), two 
populations (NAL and BB) mono cell nymphs of star gall makers (Table 2-2, 
contrast B3), and two populations (NAL and BB) of side cell nymphs from 
glabrous nipple galls (Table 2-2, contrast B2), for the same reason. There were 
six populations in all after pooling (Table 2-3). 
Within both the glabrous and the hairy nipple gall samples, there are strong 
frequency differences between side cell nymphs and center cell plus mono cell 
nymphs (Table 2-4, contrast Cl-C2). The most pronounced differences are at the 
54 
malic enzyme locus (Table 2-3). 
There are also marked differences in frequency between the side cell 
samples from the hairy nipple gall and sympatric mono cell blister gall (Table 2-4, 
contrast Dl) and between the side cell individuals from the glabrous nipple gall 
and sympatric mono cell star gall (Table 2-4, contrast D2). Thus, the side cell 
individuals are not likely to be conspecific with either of these co-occurring gall 
types. 
Allele frequencies in side cell individuals from the two nipple gall types are 
significantly different from each other (Table 2-4, contrast El). However, these 
samples are from populations -85 km apart, in very different habitats; the degree 
of differentiation between them is typical of the differences within gall morphs 
between the same two localities. A distance Wagner tree based on Rogers' 
distance for the six pooled populations grouped the two side cell taxa from 
different nipple galls together, rather than with the center cell populations from 
the same gall (Fig. 2-la). However, UPGMA analyses did not group the side cell 
samples taxa with each other or with center cell samples from their respective gall 
types (Fig. 2-lb; see also Chapter III). 
Single female rearing experiment 
None of the offspring of the eighteen green abdomen females formed 
galls, whereas offspring of most brown females caged alone did form galls (Table 
2-5). No galls were found in the control bags. The progeny of brown females 
55 
caged alone produced only single cell galls. When brown and green abdomen 
females were caged together, both single and multiple cell galls were formed, 
except that in the two cages in which the green female died early in the 
experiment, only single cell galls were formed. The results strongly suggest that 
the side cell individual is an obligatory inquiline which alone could not induce a 
gall. 
Destrnction experiment 
Some of the treated leaves were dropped before the last examination, 
reducing the number of observations in each category at the last stage from 16-30 
to 3-13. While galls in which the center nymph was left intact until the third stage 
mostly completed normal development (>91%, n=23), destruction of the center 
nymph at any of the three stages resulted in significantly smaller galls than the 
corresponding control (p < 0.01, t-test). Gall size is plotted against observation 
date for each treatment in Figure 2-2. While one may argue that the differences 
in gall size were due to gall damage when the center individual was destroyed, this 
explanation is not valid in the cases where the center individual was destroyed at 
the first stage before it was enclosed by the gall. 
None of the inhabitants of the galls that had their center cell destroyed at 
the first stage survived while their undisturbed controls developed until emergence 
(Figure 2-2 a). Leaves bearing the controls of the center cell nymphs that were 
destroyed at the first stage dropped from the tree before the last examination was 
56 
,1: 
·:: 
Ii 
performed. In one case a dropped leaf was found on the ground. No 
measurement of this first control group in the fourth time period was taken (Fig. 
2-2 a). However, the leaf found on the ground had three galls and all had obvious 
psyllid emergence holes (1-3 holes each gall), suggesting that the insects survived 
until adulthood. Those galls with center cell individuals destroyed at the second 
and third stages kept developing until the adult stage, though the survivorship was 
lower (42.8% and 55.6%, respectively) than in their control (86.7% and 100%, 
respectively) (Figure 2-2, b and c). 
Timing of female appearance 
Leaf gall maker females were occasionally found in early April, but did not 
become abundant until mid April. On April 15, 1993, none of the females caught 
in the field had green abdomens (Figure 2-3). Ten days later (April 25), the 
proportion of green abdomen females increased to 10% and in the first two weeks 
of May, it rose and remained around 70%. 
Morphological comparison 
No clear distinctions in morphology were found between center and side 
cell individuals, apart from the initial differences observed in coloration of the 
female abdomen and wing pads of 5th instar nymphs. 
57 
DISCUSSION AND CONCLUSIONS 
Taken together, the results strongly support the hypothesis that, at least 
within nipple galls, side cell individuals represent separate species from central 
cell and mono cell nymphs. The allozyme frequency differences are dramatic, 
though not absolute, and rule out free interbreeding. They are supported by fixed 
differences in coloration in two life stages, and by life history differences in female 
flight time and gall induction ability. Strong allozyme frequency and coloration 
differences between side cell samples and sympatric blister or star gallers also 
permit rejection of Moser's (1965) hypothesis that side cells are "marginal" galls 
overgrown by a nipple gall. 
Grouping of the side cell inquiline populations from different nipple gall 
types in the distance Wagner tree is consistent with the hypothesis that these 
represent a single inquiline species. However, this finding is sensitive to choice 
of clustering method and the question needs further study (see Chapter III). 
The results of the rearing and destruction experiments fully support 
Riemann's hypothesis that the side cell individuals are true inquilines. Galls were 
never formed independently by progeny of side cell females, and side cell nymphs 
rarely survived in the field if the center cell nymph was removed before the gall 
was well established. Conversely, galls with side cells were never formed by 
progenies of center cell females alone; the nymphal habit of feeding next to a gall 
inducer is presumably unique to the side cell form. As would be expected if the 
inquiline nymphs need to feed close to already-initiated galls, the phenological 
58 
observations also suggests that the side cell females come back to the buds to 
mate and lay eggs later, on average, than brown abdomen females (Fig. 2-3). 
While side cell individuals cannot initiate a gall, whether they subsequently 
contribute to gall development is not clear. In cynipid galls, the species-specific 
morphology is sometimes altered by the presence of inquilines (Evans, 1965; 
Askew, 1984; Meyer, 1987). Although one cannot in general be sure whether 
there are side cell individuals in a Pachypsylla leaf gall without dissection, it is 
sometimes possible to distinguish mono-cell galls from multiple cell galls by 
appearance. Especially in the glabrous nipple gall, the mono-cell gall is usually 
perfectly rounded, while protrusions on the gall margin are sometimes evident, 
though not always, when side cells are present. This observation, and the fact that 
the inquilines are enclosed in separate cells, suggests that the side cell individual 
can modify gall development to some degree. 
From what ancestral habit did Pachypsylla inquilinism evolve? The 
inquiline falls well within the leaf gall makers (Fig. 2-1, see also Chapter III), but 
its exact phylogenetic position is not certain - the differences among leaf gallers 
are small, with no fixed allelic differences, thus phylogenetic relationships at this 
level are not well resolved. However, it seems quite clear that the inquiline is 
derived from an ancestor which made an enclosed gall. The very few other gall 
inquilines reported in Homoptera also seem to be derived from gall formers 
(Akimoto, 1988). The same is true for at least some inquilines in other groups, 
such as cynipid wasps and cecidomyiid flies (Askew, 1984; Ronquist, 1994). 
59 
,..... 
---- --- ------
However, it is not yet clear whether all gall inquilines have this origin. 
The side cell population represents one of the closest ecological 
interactions with a congeneric gall-inhabiting species and may offer the 
opportunity to study the early stages in the evolution of inquilinism. The 
implication of this discovery for the attempt to delimit species in Pachypsylla is 
that one cannot assume, without inspecting the nymphs, that adults emerging from 
the same gall are conspecific. This may help explain part of the lasting confusion 
about species recognition in this genus. 
The fitness effects of the Pachypsylla inquiline have not yet been measured, 
but anecdotal evidence suggests a detrimental effect. Riemann (1961) reported 
that the center cell nymph can be killed by expansion of the side cells. 
How common is obligate inquilinism and under what circumstances is the 
shift to this habit favored? In Homoptera, it appears to be rare, though if 
homopteran inquilines are typically as similar to their hosts as in Pachypsylla, there 
could be more undiscovered cases. Apart from Pachypsylla, the only obligate 
inquiline that I am aware of is the aphid Eriosoma yangi, although a number of 
aphids seem to be facultatively inquiline, in that fundatrices will invade and take 
over each others' galls, particularly in cases where the gall forms at some distance 
from the site of induction. Inquilinism may be substantially more common in 
some other galling groups. In cynipids, for example, inquilinism seems to have 
multiple origins and one tribe Aulacini is dominated by this habit, though the 
monophyly of this tribe has been questioned (Askew, 1984 ). Recent studies by 
60 
Ronquist (1994) supported the hypothesis of the monophyletic origin of cynipid 
inquilines. 
The adaptive origins of inquilinism are unclear. The scenario of the 
evolution of inquilinism by host shift in Eriosoma (Akimoto, 1988) seems doubtful 
for Pachypsylla, given that the inquiline seems to have a broader host plant range 
than the individual gall forming species. An alternative explanation, not invoking 
unique historical circumstances, is that inquilinism represents a strategy for 
avoiding high mortality risk associated with gall initiation. Causes for this might 
include gall failure due to phenological mismatch with host development, and 
predation or desiccation during a longer period of exposed feeding. This 
hypothesis remains to be tested. 
61 
Chapter III. 
Phylogeny and the evolution of galling position in Pachypsylla 
62 
,,, 
:;:\ 
11:1 
,, 
·1' 
:.:1 
·),, 
.,, 
INTRODUCTION 
In this chapter, I examine phylogenetic relationships within Pachypsylla, 
using electrophoretic and morphological characters. Representatives of the two 
other genera in the tribe Pachypsyllini are used as outgroups. 
A phylogeny is necessary for any attempt to understand the evolutionary 
origins of galling diversity in Pachypsylla. Several questions from previous 
chapters are examined here. The hypothesis that the various leaf gall types are 
separate species (Chapter I), supported by significant frequency differences among 
sympatric samples, would gain support if geographically separate populations of 
the same morph were grouped together in a phylogeny. The hypothesis that side 
cell individuals from different leaf gall types in different localities represent a 
single origin of inquilinism, perhaps a single species (Chapter II), would likewise 
be strengthened if these populations grouped together. A phylogeny can also help 
answer such questions as whether the inquiline arose from a galling ancestor, and 
whether its origin preceded, and could hence have helped bring about, the 
differentiation of leaf gall morphs. 
A fourth question which can be addressed by a phylogeny concerns the 
evolution of the range of plant parts attacked by different Pachypsylla species, the 
greatest found in any group of closely related psyllids. What was the ancestral 
galling position, and did the others arise by stepwise shifts to physically adjacent 
or similar niches? Within an analogous complex of gall forming sawflies in the 
subfamily Nematinae (Tenthredinidae), it has been suggested that gall formers are 
63 
derived from a stock of free-living sawflies with a probable phylogenetic trend 
from free feeding, to leaf folding, to leaf galling to petiole and bud galling, to 
shoot galling (Smith, 1970; Price, 1988, 1992). The hypothesis that gall position 
in Pachypsylla follows a similar trend predicts that the leaf gallers should be 
phylogenetically basal, the petiole galler branches off next, and the bud and twig 
gallers form the most recent pair of sister groups, as suggested by their 
morphological and life history similarities. 
BACKGROUND 
In addition to allozyme and morphological characters, these studies 
incorporated information on chromosome number reported by Riemann (1966) 
for various species of Pachypsylla, including P. celtidisasterisca, P. celtidispubescens, 
P. celtidismamma, P. celtidisvesicula and five undetermined species, and 
Tetragonocephala flava. Riemann found that most species of Pachypsylla (leaf gall 
makers and twig galler) and T. flava have 2N chromosome numbers of 25 d and 
26 9, and suggested that these represented the ancestral chromosome numbers in 
these 2 related genera. The lower chromosome numbers in the petiole galler P. 
venusta (2N = 23 d, 24 9) and the bud galler P. celtidisgemma (2N = 22 d , 22 9) 
are derived, presumably by chromosome fusion. He further argued that the XY 
sex mechanism of the latter species was derived from an XO system in an 
ancestral Pachypsylla. The large size of the sex chromosomes suggests that this 
occurred as the result of the X chromosome fusing with one of the larger 
64 
autosomes. In this case the combined X and autosome would become the new X 
while the homologous autosome would become a Y chromosome. 
MATERIAL AND METHODS 
Samples and electrophoretic methods: 
As described in Chapter I, twenty populations (after pooling) of nymphal 
samples of Pachypsylla were included as ingroups (Table 1-6). 
A collecting trip to Asia was made in the summer of 1992 to obtain 
specimens of the outgroup Celtisaspis, purported to be closely related to 
Pachypsylla. Adults and last instar nymphs of the lerp-forrning Celtisa.\pis 
beijingana were collected by me and F. Li in June and July 1992 from Beijing and 
Shengyang, China. Only the first ins tar nymphs of Celtisaspis japonica were found 
in Seoul and Daegu, South Korea in August 1992. Since adults and the last instar 
nymphs were used for electrophoretic analyses in Pachypsylla, the first instar 
nymphal samples from Korea were not included in the electrophoretic analyses. 
Only the population of Celtisaspis from China was used. Tetragonocephala flava 
was collected by C. von Dohlen from Florida Canyon, Arizona in the June of 
1993. A total of 22 populations of nymphal samples were then analyzed, including 
20 ingroups and 2 outgroups. 
Electrophoretic methods were those described in Chapter I. Three of the 
15 loci were monomorphic inPachypsylla, but showed differences in the outgroups. 
The other 12 showed differences among gall types. 
65 
Morphological methods: 
Characters were observed both under optical and scanning electronic 
microscopy. The methods were described in Chapter I. Characters were coded 
numerically. Terminology followed Brown and Hodkinson (1988) in adult 
morphology and White and Hodkinson (1982) for nymphal morphology. 
Fourteen tax.a were analyzed, including populations from all four gall 
position groups and two outgroups (Table 3-1), and from a variety of geographic 
localities (see Table 1-3). These included eight leaf gall populations, a petiole gall 
population, two bud gall populations ( one each of the glabrous and hairy 
morphs), and a twig gall population. The leaf gall populations included two types 
of leaf hairy nipple gallers and their inquilines, the leaf star galler, the leaf disc 
galler, and two leaf blister gallers. 
Thirty-three morphological characters (Table 3-1 and 3-2, characters 1-33) 
were coded, from both adults and fifth instar nymphs. There are seven head 
characters, six thorax characters, five and four male and female genitalic 
characters, respectively. Eleven characters were coded from last instar nymphs. 
Among the thirty-three characters, four had multiple states, while the others were 
binary. 
Many possible characters were examined but rejected because of highly 
overlapping variation among tax.a. For example, the presence of caudal spurs is 
an apomorphic character inPachypsylla, but there is considerable variation in their 
number and arrangement (Figure 3-10), and in many instances there are 
66 
differences in these structures between sides of the same specimen. Therefore 
, 
this and similar characters were excluded from my analysis. 
Chromosome data 
Male and female chromosome numbers from Riemann's reports were 
coded as separate characters (Table 3-1, char. 49-50; Table 3-2, char. 131-132). 
Both characters were assigned three states, coded as the 2N number minus 20, e.g., 
the condition 2N =25 is coded as state "5." 
Life history data 
Three characters encode life history information (Table 3-1, char. 51-52; 
Table 3-2, char. 133-135). These include number of generations per year (state 
1 = univoltine, state 2 = bivoltine or multivoltine), overwintering stage (state O 
= egg, state 1 = nymph, state 2 = adult), and female oviposition time relative to 
plant development in the spring (state O = in very early spring before bud 
swelling, as in the petiole gall maker; state 1 = between bud swelling and full leaf 
extension, as in the leaf gall makers; and state 2 = after the new year's twigs are 
formed - the twig and bud gall makers lay eggs on these twigs or their axillary 
buds.) 
Phylogenetic analyses 
(1) Electrophoretic data: 
67 
Phylogeny estimation from the electrophoretic data used both distance and 
discrete character-based methods. Distance analyses employed the distance 
Wagner and UPGMA routines in Biosys-1 (Swofford and Selander, 1981) using 
Rogers' distances for both algorithms as well as Nei's distance for UPGMA. 
For discrete character analysis, the first approach treated each locus as a 
character and the alleles as unordered states (Table 3-1, char. 34-48) (Mickevich 
and Mitter, 1981). For polymorphic loci, all alleles with frequencies of 0.2 or 
more were assigned as states of that character (locus). When there was more than 
one such allele, the character was coded as polymorphic (ambiguous) (Table 3-1, 
char. 34-48). The second approach treated each allele as an independent 
character scored as present or absent (Table 3-2, char 34-130) (Mickevich and 
Johnson, 1976). An allele was scored as present in a population if its frequency 
Was 0.05 or greater. A drawback to this coding is that sampling error can result 
in false scoring. A sample of 30 individuals is necessary to detect an allele with 
frequency above 0.05 with a probability greater than 0.95 (Swofford and Berlocher, 
1987). Only five samples in the data are this large. However, the most variable 
group is the leaf gall makers, in which sample sizes are relatively large. 
(2) Morphological data: 
Most parsimonious trees for discrete allozyme characters, morphology and 
combined characters were generated using PAUP 3.1.1 (Swofford). The random 
addition sequence option for heuristic search was used in Paup, with replications 
68 
set at 1000. Analyses were performed both on equally weighted characters, and 
following the successive weighting of Farris (1988), using the rescaled consistency 
index. 
Unrooted trees were rooted using C beijingana and T. flava as the 
outgroups. A strict consensus tree was calculated when more than one most 
Parsimonious tree was found. Evolution of individual characters was analyzed 
using MacClade (ver. 3, Maddison and Maddison, 1992). 
In the analysis of morphological data, multistate characters were treated 
both as entirely ordered, and with orderings assigned to four characters (#s 1, 3, 
24
• and 27) in which a morphocline seemed plausible on inspection. 
(3) Combined evidence from morphology, allozyme, karyotype and life history: 
While twenty-two populations were scored for allozymes, only 14 
Populations were scored for morphology. When combining these two data sets, 
fourteen populations, matched in geographic region to the morphology samples, 
Were ext 
racted from the allozyme data set. 
The first approach combined morphological data, allozyme data (with each 
locus as single character), and the two karyotypic and three life history characters, 
With all characters unordered. The second approach is similar except that each 
electrophoretic allele was treated as a separate character, as described above. 
Both these analyses were also carried out with the karyotype character treated as 
Ord h d . . 
erect, that is, with the bud and petiole gallers assumed to s are a re uct10n rn 
chrom 
osome number. 
69 
RESULTS 
Pachypsylla phylogeny: 
(1) Trees based on electrophoretic data: 
No matter which coding methods were employed (frequencies or discrete), 
which genetic distance measures were applied to the frequency data (Nei's or 
Rogers'), and which grouping methods were used for genetic distances (UPGMA 
or Distance Wagner), the multiple populations within each galling position group 
were always joined together (Fig. 3-2 to 3-5). The results confirm that the leaf 
gall makers, the petiole gall makers, the bud gall makers and the twig gall makers 
each form a monophyletic group. 
Parsimony analysis treating loci as characters yielded unresolved 
relationships among the four major gall position groups. Under the other two 
phylogenetic approaches, distance methods using Rogers' distance and parsimony 
analysis with alleles as characters, the leaf gall makers are the sister group of the 
other three, and among the latter, the twig galler and the bud galler together are 
the sister group to the petiole galler. The UPGMA phenograms give a somewhat 
different picture, with the petiole galler as basal, and the outgroup 
Tetragonocephala falling within Pachypsylla, adjacent to the leaf gallers. 
Among the leaf gall makers, the genetic distances are very small (Nei's 
distances < 0.1), which suggests that the leaf gall makers are recently diverged. 
Relationships within the leaf gallers vary considerably with the type of analysis and 
cannot be considered well resolved. When loci are used as discrete characters, the 
70 
leaf gall makers are completely umesolved. The strongest suggestion of a 
grouping is the clustering of the blister gallers in both the distance Wagner and 
UPGMA analyses (Fig. 3-2 and 3-3). This does not bold, however, when alleles 
are treated as independent presence/absence characters. Similarly, the inquilines 
from the two types of nipple galls are joined together in most analyses, but not 
under UPGMA on Nei's distance. 
(2) Cladograms based on morphological data: 
Three equally parsimonious trees with a length of 45 steps (Fig. 3-6 a-c) 
were found when all characters were treated as unordered and equally weighted 
(CI=0.91, RI=0.94). Successive weighting yielded the same results. These trees 
differ only in an exchange of positions between the leaf star and blister gallers, 
and resolution of the latter. The strict consensus tree (Fig. 3-6 d) shows 
umesolved nodes within the leaf gall makers, the only groups recognized being the 
two inquilines from nipple galls, and these plus the nipple and disc gallers. 
Treating the four morphocline characters as ordered also resulted in three 
trees (length 49, CI= 0.84, RI= 0.90) and no change of the topology after 
successive weighting (Fig. 3-7 a-c). Two of these trees were the same as those 
found with all characters unordered (Fig. 3-7 a and b ). The other differs in 
grouping the petiole gall maker with the leaf gallers (Fig. 3-7 c ). 
As in the electrophoretic data, populations of the same gall position were 
always grouped together. The bud and twig gall makers were always grouped 
71 
together. The petiole gall maker is grouped either with the twig and bud gallers, 
or with the leaf gall makers. Two potential apomorphies, which are unambiguous 
only if the characters are ordered by morphoclines, place the petiole galler with 
the leaf gallers. One is surface texture of the vertex with deep wrinkles and short 
setae ( character 1, Fig. 3-8 a and b ), and the other is surface texture on the genae 
bumpy with deep wrinkles ( character 3, Fig. 3-8). The petiole gallers share four 
potential apomorphies (in characters 11, 24, 26 and 31) with the bud and twig gall 
makers, but all are ambiguous, whether or not morphoclines are used. 
(3) Trees based on combined evidence from morphology, allozymes, karyotype and 
life history: 
Ten equally parsimonious trees (length=99, CI=0.93, RI=0.93) were found 
for the combined data under locus-as-character coding for allozymes with all 
characters unordered and equally weighted. Five trees are found (length= 190, 
CI= 0.74, RI= 0.77) when alleles-as-characters coding is used instead. These 
results are not changed when chromosome number is treated as an ordered 
character. The strict consensuses for these two sets of trees differ from each other 
only in that the second is more resolved (3-9 a), and differ little from the 
consensuses for the corresponding analyses of either allozymes or morphology 
alone (Figs. 3-4, 3-5, 3-6d). In both, the four gall positions are monophyletic, the 
bud and twig gallers are sister species, and the inquilines are both monophyletic 
and related more closely to the nipple and disc gallers than to the other leaf 
72 
gallers. 
DISCUSSION AND CONCLUSIONS 
Monophyly and diversity of gall position groups 
The results strongly support the monophyly of gall position groups. In all 
analyses, all populations with the same galling positions were grouped together. 
Diversity within the twig, petiole and bud gallers is low. Petiole gall 
populations from Ce/tis occidentalis (Maryland) and C. laevigata (Texas) are 
grouped to the exclusion of that from C. reticulata (Arizona) in analyses of the 
electrophoretic data that resolve them, but the frequency differences among these 
are small (see Chapter I). Allozyme variation among geographic populations of 
bud gallers is non-significant, and the glabrous bud gallers are not grouped to the 
exclusion of the hairy bud galler, as might have been expected if these are 
different species (see Chapter I). 
Character variation among populations of leaf gallers is somewhat greater, 
but there is little clear phylogenetic information and the analysis permits few 
inferences about the species status or relationships among the gall forming 
morphs. Several analyses group the blister gall populations together, supporting 
the hypothesis that these constitute a distinct species (see Chapter I), but this does 
not always happen (Fig. 3-5). The two populations of the glabrous nipple galler, 
the other morph represented by more than one sample, are never grouped 
together. Relationships among different morphs vary considerably with data set 
73 
and analytical method. 
Perhaps the best-supported conclusion within the leaf gall complex is the 
grouping of the inquilines from two types of nipple galls, seen in all phylogenetic 
(as opposed to phenetic) analyses of both allozymes and morphology. This finding 
is at least consistent with there being only a single inquiline species. The position 
of this inquiline with respect to the true gall formers is unsettled, but it is very 
clearly inside the leaf galler complex, and undoubtedly evolved from an ancestor 
that formed an enclosed gall like that of all other extant Pachypsylla species. A 
similar conclusion, of monophyletic origin from a galling ancestor, was reached in 
a recent phylogenetic study of 
a set of cynipid inquilines (Ronquist, 1994). 
Evolution of gall position 
The phylogenetic results are consistent with, though they do not 
conclusively support, the hypothesis that evolution shifts in galling position have 
proceeded between adjacent plant tissues, i.e., from leaf to petiole to bud to twig. 
In all analyses in which the positions of the twig and bud galler are resolved, these 
species are sister groups as the hypothesis predicts (despite the contradictory 
evidence from chromosome number noted by Riemann [1961]). The petiole galler 
is ambiguously placed in some analyses, grouping with either the leaf galler or the 
bud and twig gallers. However, the latter placement, predicted by the hypothesis, 
is always favored in analyses yielding a single resolution. Finally, given the results 
74 
of this analysis, plus the fact that all other members of tribe Pachypsyllini ( and 
indeed most psyllids) are leaf feeders, the ancestral galling site in Pachypsylla is 
almost certainly the leaf blade. On any of the trees with full resolution among the 
four g 11· l d. · · a mg positions, assignment of any other ancestra con 1t1on requires an 
Instance of parallel evolution in this trait. 
75 
Chapter IV. 
Phylogenetic relationships of Pachypsyllini in Spondyliaspidinae 
and evolution of lerp and gall formation in Spondyliaspidinae 
76 
1NTRonucTION 
This study presents the first quantitative cladistic analyses of psyllid 
phylogeny. Although some Hennigian cladograms have been published (e.g. 
Bollis, 1976, 1987; Brown and Hodkinson, 1988; Hollis and Broomfield, 1989), 
these analyses have been limited in scope. The most comprehensive phylogenetic 
hypothesis yet advanced is that of White and Hodkinson (1985), which included 
all of the major groups of the Psylloidea. However, since their paper did not use 
current computer algorithms and did not present a data matrix, there remains 
confusion a d 'f' · · h · As a n controversy over phylogeny and class1 1cat10n mt 1s group. 
start toward an intended long-term re-analysis of psyllid phylogeny, this study 
focuses on the systematics of Spondyliaspidinae, with special reference to 
PachYPsyllini. 
There are three objectives in this paper. The first is to test the monophyly 
of PachYPsyllini and to estimate the phylogenetic relationships of its constituent 
genera. The second objective is to test hypotheses on the placement of 
PachYPs 11· · · 1· 'd ' G · 1 d d · Wh1'te Y 1ru within the subfamily Spondy 1asp1 mae. enera me u e m 
and Bodkinson's cladogram and related controversial taxa are sampled. The third 
goal is to look for phylogenetic trends of gall and 1erp formation in 
Spondyliaspidinae by mapping these traits on the phylogeny. Most members of 
th' . 
is subfamily exhibit one or both of these forms of concealment m the nymphal 
stage, 
77 
'I 
? 
I 
,I 
. 
BACKGROUND 
Systematics of Pachypsyllini 
Pachypsylla and its relatives Celtisaspis and Tetragonocephala comprise the 
tribe Pachypsyllini of the Spondyliaspidinae. The monotypic genus Uhleria was 
included in Pachypsyllini by Crawford (1914). However, the description of its type 
species Uhleria mira was based on a single adult specimen for which both genitalia 
and locality data were missing. Heslop-Harrison (1954) believed it to be 
introduced from Australia but offered no opinion on its placement. Taylor (1960) 
treated it as a synonym of the Australian genus Lasiopsylla, a member of 
Spondyliaspidini in Spondyliaspidinae. This species was not included in my study. 
Analysis of the works of previous authors suggests that the following 
characters define the Pachypsyllini: (1) vertex of adults quadrate, flat and large, 
(2) adult head very strongly deflexed, (3) adult pronotum vertical, ( 4) dorsal 
surface of abdomen lacking distinct sclerites in the nymph, (5) circum-anal pore 
ring of nymph absent (Crawford, 1914; Tuthill, 1943; White and Hodkinson, 1985). 
However, some of these characters are also shared by other taxa in the 
Spondyliaspidinae. Therefore, it was not clear initially which characters are 
synapomorphies for Pachypsyllini. 
All three genera of Pachypsyllini feed on Celtis subgenus Euceltis, and all 
are concealed feeders in the nymphal stages. Background information on the 
classification of Pachypsylla was given in Chapter I. The systematic and biological 
background on Tetragonocephala and Celtisaspis is given here. 
78 
(l) Tetragonocephala 
Crawford (1914) described the new genus Tatragonocephala to include the 
singles . 
pec1es T. flava, based on a single "male" specimen collected at Brownsville 
, 
Texas. Although the species has been collected in many locations, it remains the 
only included species of Tetragonocephala. He placed this genus together with 
Pachypsyfla and Uh!eria in the tribe Pachypsyllini of the Psyllinae. Tuthill (1943) 
agreed that Tetragonocephala is closely related to Pachypsylla although it lacks one 
of the essential characters of Psyllinae, i.e., spines on the basal segment of the 
metatarsus. He described more specimens from Texas, Arizona, and from Mexico. 
Tuth"l · 11 pomted out that the type specimen designated by Crawford was a female 
rather than a male and designated a male from Brownsville, Texas, as the allotype. 
The reason a psyllid expert like Crawford could make such a mistake is 
that the species has bizarre female genitalia. Unlike most other psyllids, T. /lava 
has a rounded dorsal valve instead of a triangular pointed one (Fig. 3-1). The 
dorsal l · d l · · d · va ve 1s exaggeratively enlarged and the cau a margm 1s covere with 
dense long hairs. 
Tuthill also attributed a nymph described by Ferris (1926) to T. flava. 
F~· . 
ns reported that the 5th instar nymph was found m small wax cells on leaves 
of Ce/tis reticulata at Marathon, Texas. No additional biological information was 
Provided until Riemann (l958) found several populations of Tetragonocephala in 
V • • 
anous parts of Texas on both Ce/tis reticulata and C laevzgata. He reported that 
it occurred in waxy cells similar to those described by Ferris. He reared adults 
79 
r 
I 
I 
from the nymphs and confirm that they belong to Tetragonocephala flava, which 
he predicted probably has more than one generation, unlike its univoltine relative 
Pachypsylla. 
In 1990, an outbreak of T. flava occurred in the southern United States on 
Celtis laevigata (R. Brown and J. Moser, personal communication). The 
population apparently has been reduced to low levels since then. In 1992, a 
population of T. flava was found in Arizona on C. reticulata. I reared material 
from this population in the lab on seedlings of C. occidentalis and C. tenuifolia at 
20-25°C and found that the generation time from egg to adult was about a month, 
that individuals have varying growth rates and that the generations overlap. They 
seemed to be able to survive even when plant quality was poor since they 
persisted even when all the leaves on the host were drying. 
(2) Celtisaspis 
The genus Celtisaspis was described by Yang and Li (1982) in China to 
accommodate seven Asian Ce/tis-feeding psyllids, two of which had been 
previously assigned to Pachypsy/la. The first member of Celtisaspis was a nymph 
described by Boselli (1929) from China. It feeds on Celtis sinensis, also a member 
of the subgenus Euceltis, and forms both lerps and galls. Miyatake (1968, 1980) 
subsequently described two species, Pachypsy/la japonica ( =Celtisaspis japonica) 
and P. usubai ( = Celtisaspis usubai), from Japan on Celtis sinensis var. japonica and 
gave detailed biological information. Kown (1983) described P. japonica from 
80 
I' I' 
11 :I, 
,! !. 
! 
Korea on th . . 
e same host. Takagi and Miyatake (1993) later recorded P. japonica 
from cl· 
e tzs bungeana jessoensis, which is also a member of the subgenus Euceltis. 
Ce!tisas · . 
ifJZS Japonica was the first lerp-forming psyllid known in Japan. It is 
biv I · 0 
tme whereas C. usubai is univoltine. Both of them overwinter in the egg 
stage. In C. japonica only the first generation forms both Ierps ( dome-shaped) and 
galls (horn-like) whereas the second generation forms lerps only. C. usubai forms 
both ro d . 
un lerps and pit galls. 
Yang and Li (1982) described five species of Celtisaspis from various 
regions f C 0 hina. Three of them, namely C. guizhouana, C. zhejiangana, C. 
liaonin . gana, form horn-like galls and lerps at the same time. Previous studies 
reported that C. sinica produces slight pit galls and white oyster lerps. C. 
beljzna . 
0 ana forms lerps only (Yang and Li, 1982) but my observat10ns suggest that 
the nymphs also cause deformation of the leaves on which they form lerps. Yang 
and Li l · · · b d 'd reported the host species of c. sinica to be Ce tzs smenszs ut 1 not 
specify h · · · · · Ch· w 1ch species of Celtis others feed on. Durmg my collectrng tnp m ma, 
Li anct I found Celtisaspis beijingana on both Celtis sinensis and C. bungeana. 
Yang and Li considered Celtisaspis to be closer to Tetragonocephala than 
Pach l 
'YPsylla since the former two genera have longer genal cones, onger antennae, 
an· . 
Ind1stinct pterostigma, a short rounded female dorsal valve and have the lerp-
for · 
ming habit. They agreed with Heslop-Harrison (1954) that all three genera 
Should be placed in Spondyliaspidinae rather than Psyllinae. 
81 
Systematics of Spondyliaspidinae 
The Spondyliaspidinae, later elevated to family rank (White and 
Hodkinson, 1985), was described by Schwarz (1898) for the Australian lerp-
forming genus Spondyliaspis but no definition of the subfamily was given. Heslop-
Harrison (1949, 1951, 1954, 1958 and 1959) reviewed the higher classification of 
world psyllids and first properly defined the Spondyliaspidinae. He distinguished 
them from the Psyllinae by the following characters: bipartite male proctiger; 
presence of anteoccipital lobes; tubular or quadrate genae that possess a single, 
large, peg-like seta; stout antennae; and infrequently the presence of meracanthal 
spurs (Heslop-Harrison, 1958). Thirteen genera were included; they are from 
Australia, New Zealand, Central and North America and the Inda-Malayan 
Archipelago (Heslop-Harrison, 1954). 
Most authors since Heslop-Harrison have recognized the group, although 
there has been some dispute on its definition and boundaries. Four subfamilies 
of Spondyliaspididae were contained in White and Hodkinson's system, namely 
Arepuniinae, Euphalerinae, Pachypsyllinae and Spondyliaspidinae, based on both 
nymphal and adult characters. However, these four subfamilies are polyphyletic 
in their phylogeny (Fig. 4-1). Taylor (1990) transferred the tribe Ctenarytainini 
from Aphalarinae to the Spondyliaspidinae based on the similarity of adult 
morphology. Burckhardt (1991) pointed out that some of the clades in White and 
Hodkinson (1985) were defined by either plesiomorphic or homoplasious 
characters and required reexamination. He defined Spondyliaspidini, included 
82 
Ctenaryta· . 1
m as a junior synonym and excluded Euphalerini from the 
Spondylias "d. 
PI mae (Table 4-1 ). His earlier paper (Burckhardt, 1987), based on the 
structur 
e of larval tarsal arolia, placed the Arepuninae outside of the 
Spondylia "d · . 
spi mae m the Aphalaroidinae. These actions left only two tribes in the 
Spondyliaspidinae, i.e. the Spondyliaspidini and the Pachypsyllini. Although 
Burckhardt (1991) defined Spondyliaspidini and provided constituent genera, he 
did not defin "th · · · h 11· · · d f" · · e e1 er Spondyhasp1dmae or Pac ypsy 1rn m a e 1mt1ve manner. 
Thus, a clear phylogenetic definition for this subfamily is needed. 
Th_e Doc,;.:, 
~on of Pachypsyllini within Spondyliaspidinae 
From host plants and biogeography, one might question whether 
PachyPsy111·m· belong in Spondyliaspidinae. All other Spondyliaspidinae are 
endemj . 
c to the southern Hemisphere, mostly Australia, and feed on Myrtaceae, 
lllostJy Eucalyptus. Members of Pachypsyllini occur in the Holarctic, and develop 
0n Celt" 
zs spp. (Ulmaceae ). 
Recent authors nonetheless agree that Pachypsyllini is correctly placed in 
Spondyliaspidinae (Klimaszewski, l964; Loginova, 1964; Becker-Migdisova, 1973; 
White d h h b an Hodkinson, 1985; Burckhardt, 1991). However, t ere as een no 
rigorou · · f 
s phylogenetic test of this proposition, and the exact pos1t10n o the 
p 
achyPsy111·m· 
m the subfamily is uncertain. White and Hodkinson regarded 
PachyPsyllini as . t t v'hel''omm!!a which is a member of their a sis er group o r, ,, rUJ , 
PoIYphyletic Euphalerinae within the Spondyliaspididae ( = Spondyliaspidinae) 
83 
sensu Wb· 
Ite and Hodkinson. In contrast, Burckbardt (1991) placed Pachypsyllini 
as the sister group of the Spondyliaspidini, in which he included Phellopsylla. 
l-Ieslop-Harrison (1949) in his discussion of the strong linkages between many 
a . 
nunal and plant groups of Australia and Southern America, stated that 
Ctena,yt . 
azna, a small genus in Australia, New Zealand and the South Pacific 
Islands "h . 
' as its nearest relatives in the anomalous genera Pachypsylla Riley, 
Tetrao0 0 
nocephala Craw. of Central and Southern North America." These 
hYPotheses are tested by reconstruction of the phylogenetic relationships of 
Pach}'ps 11· . 
Y 1ru and other Spondyliaspidinae in this study. 
~feeding (lerp- and gall-forming) in Spondyliaspidinae 
The Pachypsyllini resembles other Spondyliaspidinae in constructing lerps 
and gaU . . 
s, Unlike nearly all other North Temperate psyllids. The majority of 
PSYllids a f . b h 
re ree-living. However, at least two types of shelter construct10n Y t e 
nYnlphaJ stages have evolved, 1erp formation and gall induction. These habits 
illustrat . d 
ea Widespread evolutionary tendency in sternorrhynchous Homoptera an 
10 
a nu b · · 1 d · b ·1 m er of other insect groups, i.e., the ongm of a concea e , 1mmo 1 e way 
Of life fr 
om more or Jess free-living forms. 
A lerp is a case constructed from a carbohydrate secretion from the anus 
of PsylJid nymphs (honeydew) which hardens upon exposure to the air (Dobson, 
1851. W . 
, hite, 1972). It has been reported that in some psyllids, e.g. Celtisaspis 
ltsuba· 1 d" h ' d wax 
z and M acrohomotoma sp., the lerp is a mixture of ana isc arge an 
84 
filaments which is similar to the test of the armoured scale Conchaspis (Takai and 
Miyatake, 1993). Each lerp is inhabited and constructed by a single nymph and 
usually occurs on the leaves. The structure of the lerp is constant within species 
(Morgan, 1984). Most lerps look rather like the shells of a bivalve but some have 
the appearance of oyster scale insects, and some form a horizontal tube. 
Lerps have generally been accorded a protective function, as a barrier 
against predator and parasite attack, a reflective shield against radiation, or a 
means of avoiding desiccation (White, 1970). Most lerp formers occur in the dry 
regions of Australia, feeding on Eucalyptus. White (1970) worked on the feeding 
biology of the lerp-forming Cardiaspina densitexta on Eucalyptus in Southern " I 
:! 
Australia, where the relative humidity frequently drops to 10-15% in the summer. 
He found that the mouthparts of the nymphs are always inserted through a stoma 
and that there is almost no air movement under the lerp. There is a remarkable 
increase in humidity from water vapor transpired by the leaf beneath this 
structure. The high humidity within the lerp increases the probability that the 
stomata beneath the lerp will stay open, while elsewhere on the leaf they remain 
closed to reduce transpiration. 
Galls are generally interpreted as abnormal growth of plant tissues caused 
by various organisms that irritate the plant and possibly lead to the production of 
a growth hormone (Meyer, 1987). Most galls of psyllids are formed by the feeding 
of the nymphs. Partial galls, such as pit galls or rolled leaf galls, can be 
distinguished from complete, enclosed galls, which are exemplified by those of 
85 
Pachypsylla species. These broad categories of gall types may represent different 
degrees of evolutionary advancement. The phylogenetic conservatism 
exemplified in the retention of ancestral concealment forms in the Pachypsyllini 
may be mirrored in the evolution of gall variation within Pachypsylla. In an 
analogous gall forming group of insects, the sawfly subfamily Nematinae, it has 
been suggested that gall formers are derived from a stock of Nematus-like free-
living sawflies, with a probable evolutionary sequence from free feeding, to leaf 
folding, to leaf galling to petiole and bud galling, to shoot galling (Smith, 1970; 
Price, 1988, 1992). Evidence consistent with a similar trend in gall position in 
Pachypsylla was presented in Chapter III. 
Hodkinson (1984) proposed that Pachypsyllini illustrate another 
evolutionary trend in the form of psyllid concealment, from lerp construction to 
gall formation as in Pachypsylla. In contrast, Moore (1970), in a phylogenetic 
study of Glycaspis (Spondyliaspidini) hypothesized a sequence in which galls came 
first then gave rise to flat lerps, to round lerps, to oval lerps, to rectangular lerps, 
and finally to no lerps. To test the existence and generality of such hypothesized 
trends, it will be necessary to identify the nearest relatives of Pachypsyllini among 
the Spondyliaspidinae. 
MATERIALS AND METHODS 
Sampling of taxa: 
Twenty-seven species in 16 genera were analyzed (Tables 4-1, 4-3 ), 
86 
i:;:. 
\· 
I' 
including all three genera of Pachypsyllini. Within Pachypsylla, four species 
representing three major galling position groups were sampled, including a leaf 
blister gall maker, P. celtidisvesicula, a hairy nipple gall maker, P. celtidismamma, 
a petiole gall maker, P. venusta, and a glabrous bud gall maker, P. cletidisgemma. 
To estimate major lineages in Spondyliaspidinae, representative genera of the 
major groups according to White and Hodkinson (1985), as well as other genera 
included in the subfamily by Burckhardt (1991), were studied (Fig 4-1, Table 4-1). 
These taxa are Creiis, Spondyliaspis, Cardiaspina, Glycaspis, Colophorina, 
Euphalerus, Phellopsylla, Retroacizzia, and Arepuna sensu White and Hodkinson 
(see Table 4-1). The trees were rooted by including representative genera of 
other subfamilies of Psyllinae, namely, Trigonon, Acizzia, Heteropsylla, Psylla and 
Russelliana. One to three species of each genus were sampled. Five to 30 
specimens, slide mounted or pinned, were examined, except that for some only 1 
to 4 specimens were available (Appendix 4-1). Phellopsylla was sampled in order 
to contrast White and Hodkinson's hypothesis that Phellopsylla, placed in their 
taxon Spondyliaspididae: Euphalarinae, is the sister group to Pachypsyllini, versus 
Burckhardt's inclusion of Phellopsylla in his Spondyliaspidini. Ctenarytainini was 
included to test White and Hodkinson's (1985) exclusion of this tribe from their 
Spondyliaspididae, versus Taylor's (1990) and Burckhardt's (1991) inclusion of it 
in that taxon. Similarly, Arepuna ( = synonym of Russeliana) is included to test 
White and Hodkinson's inclusion of this genus in Spondyliaspidini, in contrast to 
Burckhardt's exclusion of the genus from that tribe. 
87 
Morphological methods: 
Adult and 5th instar nymphal morphological characters were examined in 
pinned and slide-mounted specimens. Specimens were obtained from material 
collected by the author and colleagues, from the psyllid collection of U.S. National 
Museum of Natural History, Smithsonian Institution (USNM) and from loaned 
specimens (Appendix 4-1). Slides were prepared from dry or deep-frozen 
specimens which were dissected under the stereomicroscope after treatment with 
KOH. Parts were slide mounted in Canada Balsam. Some slides borrowed from 
the Australian National Insect Collection were originally mounted in polyvinyl 
lactophenol (see Appendix 4-1). 
Characters that were used to define groups of Spondyliaspidinae in White 
and Hodkinson's (1985) cladogram were carefully examined. Most were included 
in my analyses; however, the treatment is different. White and Hodkinson tend 
to treat what are logically the alternative states of a single character as 
independent, presence/absence characters. For instance, their character 71 is 
"pterostigma reduced" while character 72 is "pterostigma absent or very reduced", 
each with two states (present and absent). The pterostigma in my analyses was 
treated as a single character (#13) with two states, "obvious" versus "absent or very 
reduced." Other similar examples include adult metatarsus segment I with one or 
no spines ( their characters 78 and 79; my character 16) and wing apex shape ( their 
characters 63-65; my character 11). 
Forty-five characters were used, including 10 adult head characters, 7 adult 
88 
I ,\ 
I 
thorax characters, 5 male genitalic characters, 4 female genitalic characters and 
19 nymphal characters (Tables 4-2, 4-3). Twenty-eight characters were binary, 
while the remaining eighteen had multiple states. 
Terminology follows Brown and Hodkinson (1988) for adult characters, and 
White and Hodkinson (1982) for nymphal characters. 
Morphological data analyses: 
Morphological characters were coded numerically. Polymorphism within 
a terminal taxon was coded as ambiguity (e.g. character 14). In an initial analysis, 
with all characters unordered and no outgroups specified, the two species of 
Russelliana grouped with the four taxa of Psyllinae. In subsequent analyses, these 
six taxa were designated as outgroups. Characters were first treated as all 
unordered and of equal weight. Successive weighting according to the rescaled 
consistency index was then applied (Farris, 1988). In a separate analysis, 
transformation series were assigned for multistate characters 3, 4, 5, 6, and 7, as 
discussed below. Equal weighting was followed by successive weighting as before. 
Transformation series were erected only for characters in which a 
morphocline seemed clearly plausible. Character 3 describes irregular depressions 
or fovea on the vertex, which are always present in psyllids (Brown and 
Hodkinson, 1988). Usually these appear as a pair, one on each side of the vertex, 
with different degree of expression in different taxa, sometimes obvious and 
sometimes not. In Trigonon longicomis, the depression appears as a vertical curve 
89 
bearing two pits or indentations on each side. In Phyllolyma th . d . 
' ree m entations 
were seen. The number of obvious indentations which varied fro 
' m zero to three 
was treated as a stepwise transformation series. 
, 
In psyllids, the genae are often swollen to form genal processes or cones 
, 
which provide useful characters (Brown and Hodkinson 1988). Character 4, which 
encodes the length of the genal cones compared to the length of the vertex, is 
treated as a linear transformation series. 
Character 5 describes variation in the shape of the occipital foramen, a 
previously overlooked feature. The apparent shape varies with the viewing angle, 
but is constant within genera when the head is positioned horizontal to the vertex. 
Three shape categories were recognized (Fig. 4-2). The apparent intermediacy of 
type b between types a and c was encoded in the transformation series a-b-c. 
The antennae of psyllids show different degrees of modification. Character 
6 encodes the length of the antennae standardized by the width of head including 
eyes, and is treated as a four-state linear transformation series ranging from 0.7 
to 2.0. 
Most psyllid antennae have ten segments. The basal two segments are 
usually the shortest and the third the longest. In Pachypsylla the tenth segment 
is the shortest. Character 7 encodes the length of antennal segment X to 
compared to that of segment I, and is treated as a linear transformation series. 
The phylogenetic significance of one character, the apical spines (saltatorial 
spurs, character 14), might be questioned. The number of these spines is variable 
90 
-
Within species, sometimes even within one individual, and there is overlap between 
species. Hollis (1984), in his revision of Afrotropical Triozidae, did not consider 
the apical tibial spurs a good phylogenetic character since they have arisen more 
than once. In contrast, Brown and Hodkinson (1988), in their study of 
Panamanian psyllids, considered the number and position of these spurs to be an 
important character in the higher taxonomy of the Psylloidea. White and 
I-Iodkinson's analyses divided the number of spines into two groups, more than 6 
or less Th · · d D · · · b'l' 
· e same method of coding was used m this stu y. esp1te 1ts vana 11ty, 
th· 
IS character had a retention index of 0.75, about average. Therefore, it was 
retained in the analysis. 
Phylogenetic reconstruction was carried out first in Hennig86 ver. 1.5 
(Farris, 1988) and then in PAUP ver 3.1.1 (Swofford, 1993) based on the 
Parsimony criterion. In PAUP, the random addition sequence option for heuristic 
search Was used, with replications set at 1000. Additional character analyses were 
Performed using MacClade ver. 3 (Maddison and Maddison, 1992). 
Life h' . . . . ~ mformation, character codmg and hxvotheses testmg. 
Relatively little attention has focused on the life history and ecology of the 
PsyIIoidea as compared to other sternorrhynchous homopterans such as aphids and 
scale insects. The relevant literature is widely scattered and has not been 
comprehensively reviewed. 1 compiled the available information on the feeding 
habits for each genus included in the phylogenetic analysis. These data are 
91 
, ,I 
11 
presented in Results. 
To reconstruct the evolutionary history of concealment forms on the 
cladogram, the variation in habits must be coded as character states. The best 
way to do this is not obvious. Therefore, two approaches were taken. The first 
coding method (Table 4-4, coding 1) is a simplified contrast among three 
conditions treated as states of a single character, namely, free-living (F or 0), lerp-
forming (Lor 1), and gall-making (G or 2). This coding is the simplest rendering 
of the hypothesis of Hodkinson described earlier, in which lerp forming is 
postulated to be intermediate between free living and full concealment as in galls. 
Species that do not form galls or lerps, even if they produce ample wax or 
flocculent, were coded as free-living, and no distinction was made between partial 
and enclosed galls. Celtisaspis beijingana, which forms both lerp and gall, was 
coded as polymorphic (ambiguous). 
The second approach coded lerp-forming and gall-making as independent 
characters, which seems appropriate because these concealment forms are 
structurally non-homologous and can be found together in the same species. For 
gall formation, it is postulated that partial gall-forming (P or 1) is phylogenetically 
intermediate between enclosed gall-making (G or 2) and no host modification (N 
or O) (Table 4-4, coding 2). For lerps, it is postulated that production of abundant 
wax and honey dew flocculence (W or 1) is an intermediate stage in the evolution 
of structured lerps (L or 2) from free living (N or 0), defined as production of 
neither lerp nor flocculence (Table 4-4, coding 3). 
92 
To test the hypotheses embodied in them, these codings were superimposed 
on the phylogeny of Figure 4-3 (tree b), inferred from morphological characters 
excluding feeding habits, to determine the minimum number of evolutionary 
changes required by each. The number of steps required under the postulated 
Ordering . 
was contrasted to that under all possible other ordering. These 
calculaf ) ions were done using PAUP ver. 3.1.1 (Swofford 1993 and MacClade ver. 
3 (Maddison and Maddison 1992). 
To test statistically whether there is any phylogenetic component to the 
distribution of concealment types, a randomization approach was used (Faith and 
Cranston 1991; Liebherr and Hajek, 1990; Maddison, 1990). The inferred number 
of independent origins of concealment types was contrasted to a null distribution 
Obt ' 
ained by 1000 random reassignments of the character states to taxa on the 
cladogram, generated by the PC Pascal program "MMMAT' (G. Roderick, 
Unpublished). The minimum required number of origins of concealment types 
(treated as unordered character states) under each randomization was calculated 
in PAUP 
RESULTS 
~logenetic relationships 
When all characters were treated as unordered, three most parsimonious 
tre d · · d es (Figure 4-3) were found, with a length of 124 steps an a consistency rn ex 
( CI) and retention index (RI) of 0.55 and o. 78, respectively. The same three trees 
93 
are found by successive weighting. The strict consensus of these, shown in Figure 
4-3 ( d), is identical to the Adams and 50% majority rule consensus trees. The 
three most parsimonious trees differ only in the way the two species of Creiis are 
related to Cardiaspina. As in the earlier analysis of Pachypsyllini (Chapter III), 
relationships among the Pachypsylla leaf gall makers are unresolved. 
When the five characters (3-7) for which morphocline transformation series 
were hypothesized were treated as ordered, 164 equally parsimonious trees with 
a length of 134 steps were found with CI = 0.51 and RI = 0.75. Among these are 
the three trees found for unordered characters. The strict consensus tree is shown 
in Figure 4-4. There is considerably less resolution than in the unordered analysis 
(17 groups resolved versus 24). Ambiguities in addition to those appearing in the 
unordered analysis include the positions of Spondyliaspis, Euphalerus sp. and 
Acizzia. Successive weighting reduced the number of trees to ten, all included in 
the original 164 but none identical to any tree found with unordered characters. 
Their strict consensus tree is shown Figure 4-5. This tree is nearly as resolved as 
that for the unordered analysis, except for the position of Spondyliaspis, and 
disagrees only slightly, in the position of Euphalerus sp. A. 
Lerp and gall forming biology in the Spondyliaspidinae 
The literature search on feeding habits produced the following summary for 
the genera included in the phylogenetic analysis. 
(1) Pachypsylla is a North American enclosed gall former, feeding exclusively 
94 
(2) 
feeding on hackberry Celtis subgenus Eucelti.s. The plant tissues attacked 
include the leaf blade, petiole, bud, and twig, each by different species 
(Osten Sacken, 1861; Riley, 1876-1890; Tuthill, 1943). 
Tetragonoceplzala is found in the southern United States and northern 
Mexico feeding on Celti.s subgenus Eucelti.s. This monotypic genus forms 
(3) 
a round white lerp on the leaves (Crawford, 1914; Riemann, 1958). 
Ce!tisaspi.s is distributed in Asia (Japan, Korea and China) and also infests 
(4) 
(5 ) 
(6) 
CI · e !ts subgenus Eucelti.s. Nymphs feed on the undersides of hackberry 
leaves, producing partia l (pit or horn-shaped) ga lls or sometimes just 
mconspicuous deformation of the leaf, and are always concealed within 
le rps (Boselli, 1929; Miyatake, 1968, 1980; Yang and Li, 1982; Kwon, 1983). 
P/zeflopsyl!a, originally described as Thea, is an Australian Eucalyptus 
fee de r. Species of Phellopsylla are found under smooth barked eucalypts, 
fee ding o n the bark and excreting long white waxy fil aments a t va rious 
levels on the main trunk, usually under pieces of dried ba rk still loose ly 
a ttach ed, up to the terminal branchlets (Taylor, 1990). 
Plzyl!olyma is a Eucalyptus feeder in Australia, forming white round lerps 
on the te rmina l branchJets or bivalve lerps inserted into the margins of til e 
leaves. T he white round Jerps are unwoven and fragile. In some cases, 
more than one species may occupy the same site (Morga n, 1984; Taylor, 
1990). 
Crehs· forms oyster-shaped 1erps on the leaves of Eucalyptus in A ustralia 
95 
(7) 
(Froggatt, 1900; Morgan, 1984 ). 
Spondy!iaspis forms brown shell-like lerps on Eucalyptus leaves in Australia 
(Morgan, 1984; Carver, 1987) 
(8) 
Cardiaspina is one of the most divergent Spondyliaspidinae (Taylor, 1960, 
l962). It is an Eucalyptus feeder. Different species form various kinds and 
colors of lerps (Cambell, 1964; Crawford, 1911; Morgan, 1984; Taylor, 
l989, 1992; White, 1970). 
(9) 
Glycaspis is a large Australian group that forms either lerps or galls on 
Eucalyptus. Moore (1970) divided the genus into three subgenera, of which 
Boreioglycaspis was raised to the generic level by Burckhardt (1991). The 
00) 
remaining subgenera are Synglycaspi.s, which builds round, oval, or 
rectangular lerps, and Glycaspis, which makes galls, flat lerps or round lerps 
(Moore 1970). 
Ctenarytaina is an Australian/ Asian/ Pacific genus. All the known species 
are free-living, feed on shoots of Eucalyptus, and produce viscous globules 
and flocculent threads. One species, C longi.cauda, was found feeding 
among young leaf buds that were still closely folded, and it exuded thin 
White threads (Morgan 1984; Taylor, 1987; Carver, 1989). 
(11) c ( 
olophorina occurs in Southern Africa on Cassia petersiana Legurninosae ). 
The nymphs are covered with fine wax particles, and live and develop 
between pairs of unopened leaflets which form a globular pouch (Capener, 
1973). 
96 
.... 
(l2) Euphalerus is a heterogeneous legume-feeding group found in both the 
New World and the Old World. Both gall-making and "nest-constructing" 
( ==lerp-forrning?) species are found (Russell, 1971). 
(l3) Russelliana ( == Arepuna sensu White and Hodkinson) is a free-living South 
American group feeding on Solanaceae (Tuthill, 1959). 
(l4) Heteropsylla is a New World legume-feeding and free-living genus 
(Crawford, 1914; Beardsley, 1987). 
(15) 
Acizzia is a free-living and legume-feeding genus typical of the warmer 
regions of the Old World, Africa, Arabia, India and Australia (Morgan, 
1984; Carver, 1987; Heslop-Harrison, 1949). 
(l6) Tngonon is a small Austral-Oriental genus with unknown feeding habits 
(Crawford, 1920). 
(l?) Psy!!a s. str. is a Holarctic genus, members of which are free-Jiving on 
Betulaceae and Carpinaceae (Crawford, 1914; Hodkinson, 1988). 
The ev I . . 
~ut1on of concealed feeding in psylhds 
In the first coding method (Table 4-4, coding 1), under which 1erps and 
galls a · · h d · t d d · re treated as part of a single transformat10n senes, t e pre 1c e or ermg, 
Rt-G, requires 4 steps, the same as does no ordering at all, while the alternatives 
RG-L and L-F-G require 7 and 5 steps, respectively. Under all three 
transformation series as well as no ordering, significantly fewer steps are required 
than for the same data randomized on the tree (p<O.OOOl). 
97 
Under the second coding, galls and lerps were treated independently. For 
galls the d. . . 
' pre 1cted ordenng, N-P-G, reqmres 3 steps, as does the alternative N-G-
P, but the alternative P-N-G requires the minimum possible, 2 steps. Thus, the 
Predictio · 
n 1s not supported. For lerps, the predicted ordering N-W-L requires 6 
steps on 
' e more than the alternative N-L-W, which requires 5. For both lerps and 
galls, significantly fewer transformations are required than under the null 
hyPoth · · 
esis of mdependence of phylogeny (randomization test, p < .05 and p < .001, 
respectively). 
D1scu 
SSION AND CONCLUSIONS 
The phylogenetic analyses confirm that the Pachypsyllini are a 
rnonoph J • h · · Y et1c group. Tetragonocephala and Celtisaspis form a clade w 1ch 1s the 
sister gr · Wh" d H dk' ' oup to Pachypsylla. This result is congruent with 1te an o mson s 
(1985) conclusion. However, the characters defining the group are somewhat 
differe 11 · · · Wh · 
nt. Three characters were presented as defining the Pachypsy 1ru m 1te 
anct Bodkinson's phylogeny, namely, anteoccipital lobes absent (their char. 53), 
rhin- · 
anum absent from seg. v (their char. 58), and dorsal surface of abdomen 
lacking d' . · h 1 · 1stmct sclerites (their char. 126). The second character 1s omop as1ous 
and Probably plesiomorphic within the Psylloidea (Burckhardt, 1991). The first 
and third characters are true synapomorphies ( characters 10 and 33 in my 
analyses). In my analysis another character was found that also defines the group, 
98 
i.e. the presence of the nymphal trochanters (character 30). 
Since the hypothesized morphoclines decreased rather than enhanced tree 
resolution, presumably because they increased character conflict, I place more 
credence in the analysis using only unordered characters. The three trees 
obtained with the latter differed only in the position of the two Creiis species; the 
monophyly of Creiis needs further testing. For comparison with White and 
Hodkinson's results, tree (b) in Figure 4-3, which grouped the two Creiis species 
together, was chosen. This tree differs from White and Hodkinson's (1985) 
cladogram in many places. The main differences are as follows: 
(1) Phellopsylla is not the sister group to Pachypsyllini in my analyses. Instead, 
it and Phyllolyma together are the sister group of the rest of 
Spondyliaspidinae sensu Burckbardt and are a group within 
Spondyliaspidinae sensu White and Hodkinson. One character uniting 
Phellopsylla and Pachypsyllini in White and Hodkinson's tree is the 
bipartite male proctiger. This character is hard to see. Burckhardt (1991) 
reported that the second segment is present in Phellopsylla, but is reduced 
and membranous. Examination of the specimens available to me yielded 
no suggestion of the presence of a second segment. Therefore, I coded the 
proctiger in these groups as unipartite. Changing the coding to bipartite, 
to agree with Burckhardt's and White and Hodkinson's codings, did not 
change the results. More observations are needed to determine which 
coding is correct. 
99 
(3) 
(2) 
Euphalerus nidifix is the sister group to all other Spondyliaspidinae sensu 
White & Hodkinson in my analysis, whereas in White and Hodkinson's 
cladogram, it is next to Phellopsylla and in the middle of the 
spondyliaspidid clade. Euphalerus was a polyphyletic taxon on White and 
Hodkinson's cladogram; different species of the genus were in different 
clades of Spondyliaspidinae. Burckhardt (1991) removed Euphalerini sensu 
White and Hodkinson from Spondyliaspidinae since it has a larval tarsal 
arolium with a pedicel and a visible unguitractor. In my analyses, the 
arolia and unguitractor are absent in all species above Colophorina and 
Euphalerus, whereas all others have them. Euphalerus sp. A sensu White 
and Hodkinson shares more character states, though plesiomorphic, with 
Co!ophorina than with E. nidifix. It is not likely that Euphalerus sp. A is 
congeneric with E. nidifix. As suggested by D. Hollis (per. comm.) and 
White and Hodkinson's tree, Euphalerus sp. A. might turn out to be a 
Sp . 
ec1es of Colophorina, although these are not always grouped together. 
Ctena,ytaina branches off above (Phellopsylla + Phyllolyma), well within the 
Spondyliaspidinae. It was excluded from the Spondyliaspidinae and placed 
in Euphyllurinae (Aphalaridae) by White and Hodkinson. This hypothesis 
is not conclusively ruled out by my study, since no other members of 
Euphyllurinae were sampled, but my results support instead the views of 
Taylor (1990) and Burckhardt (1991). Taylor (1990) transferred 
Ctenarytainini to the Spondyliaspidinae based on the similarity of adult 
100 
/'I' 
1,11; ,,., 
·I /ii,· 
If /:/ 
I II f 
;! !\/ ;I , , 
'u 
;; /·11 
,. I~ 
.I '.;, 
/1 ; I jl '' 
J, ' ! 
I• J •I 
II I I JI 
' ' 
characters with other spondyliaspidines. He also pointed out that the only 
significant differences between Ctenarytainini and Spondyliaspidini are the 
presence of a caudal plate and lanceolate setae in the larvae. Burckhardt 
(1991) further noted that Ctenarytainini have the spondyliaspidine nymphal 
aroliurni, i.e., membranous without pedicel and visible unguitractor, and 
rejected the free-living habit of Ctenarytainini as grounds for removal from 
Spondyliaspidini, since concealment habits are quite homoplasious. 
( 4) The relationships within the Creiis + Cardiaspina + Glycaspis + 
Spondyliaspis clade are completely different between my tree and White 
and Hodkinson's. 
My results also cast doubt on the monophyly of Spondyliaspidini sensu 
Burckhardt, part of which forms the sister group to Pachypsyllini. Further 
sampling of Spondyliaspidini and Ctenarytainini sensu White and Hodkinson and 
Taylor is needed to test this finding. 
Heslop-Harrison (1949) suggested that the Australian psyllids originated in 
part, during early Tertiary times, as immigrants from Antarctica, and in part 
through. He assigned psyllid fossils of that age from Tasmania, presumably a 
major corridor between these two continents, to the genus Ctenarytaina Ferris and 
Klyver. The relatively basal position of Ctenarytaina in my cladogram of the 
Spondyliaspididae, a largely Australian group, is consistent with Heslop-Harrison's 
view. 
101 
The evolution of concealed feeding in psyllids 
The analysis of concealment habits under coding method #1 supported 
Hodkinson's postulate that lerps tend to precede galls in spondyliaspidine 
phylogeny. Within galls and lerps considered separately, however, there is no 
support for the idea that partial galls or lerps represent a phylogenetically 
transitional stage from free living. These tests, it must be stressed, are heuristic 
only. No statistical test was applied, and the data are very incomplete. Outgroup 
sampling and life history information 1s sparse, and even within 
Spondyliaspidinae, numerous genera and species are unrepresented in the analysis. 
Despite limited data, however, there is strong support for the more general 
postulate that concealment forms characterize phylogenetic groups. Thus, we can 
reject the alternative hypothesis that these features reflect only rapid and variable 
adaptation to local environmental conditions, appearing and disappearing with no 
phylogenetic pattern. It will therefore be of interest to look further into possible 
evolutionary trends in, and consequences of, those habits. 
In a survey of gall-inducing families among arthropods, Roskam (1992) 
distinguished two patterns in the taxonomic distribution and host associations of 
gallers. In major groups (families) of gall inducers, e.g. Cecidomyiidae, most 
members are gall makers and their host plants include diverse angiosperms. 
These are likely to have evolved from an ancestor that already possessed the gall-
inducing ability in an early period of angiosperm radiation. In contrast, minor 
galling groups, such as Curculionidae, Tenthredinidae, Cephidae, Gelechiidae, 
102 
Tephritidae, and Agromyiidae, have only a few exceptional members making galls. 
These represent la te radiations confined to particular plant genera. In Roskam's 
view, homopteran and thrips gall inducers are intermediate between these two 
groups. The numerous galling members are scattered among non-galling ones and 
across many plant groups. His interpretation is that sap-feeding offers 
homopterans considerable opportunities for manipulating host plant; as a 
consequence, gall inducers evolved independently many times from sap feeders, 
over a long geological time span. 
Roskam's explanation seems plausible. Galling psyllids occur in various 
clacies. The phylogenetic pattern may not be demonstrated at higher levels but 
is evident in closely related groups. Burckhardt (1991) suggested tl1 at 
diversification of Spondyliaspidini may have occurred, together with th at of their 
myrtaceous hosts, during the mid-Eocene. The earliest fossil Psylloidea come 
from the middle (Handlirsch, 1925; Klimaszewski , 1993a) and upper Jurass ic 
per iods (Becker-Migdisova, 1949, 1985). If this is true, Spondyliaspidini is 
re la tively young compared to other major galling insect groups and other psyllid 
clades. By the O ligocene the eucalypts and acacias became importan t elements 
in the vegetati on of Australia, which became increasingly arid. Fire resistance 
probably played an important role in the high diversity of present-day Australian 
Myrtaceae. In conjunction with host-plant speciation, development of lerp and 
ga ll concealment might have furthered spondyliaspidine diversification by 
protecting them from environmental extremes. 
103 
There are many hypotheses on the adaptive value of gall formation. Price 
et al. (1986, 1987) recognized six explanations. These include the "nonadaptive" 
hypothesis, the "mutual benefit" hypothesis, the "plant protection" hypothesis, the 
"nutrition" hypothesis, the "microenvironment" hypothesis, and protection from 
enemies. They rejected the first three hypotheses since galls clearly benefit the 
gall forming insect, not the plant. 
The relative importance of these alternative effects, for psyllid gallers, is 
unclear. Many studies have suggested that galls do not protect psyllids from 
natural enemies (Smith, 1970; Askew, 1980; Washburn & Cornell, 1981). They 
may even increase susceptibility to enemies which seek out aggregations of hosts. 
The "microenvironment" hypothesis, that galls serve to protect psyllids from 
climatic extremes such as desiccation, seems plausible, in the absence of evidence 
to the contrary. There is also evidence consistent with the hypothesis that gall 
formation improves psyllid nutrition. It has been found that in the nipple galls 
formed by Pachypsylla celtidismamma on Celtis, lipid material is abundant basal 
to the central depression, and carbohydrates are also present (Beisler, 1989; 
Beisler and Baker, 1992). However, the importance of these supplies to the 
nymphs has not been studied. 
Heslop-Harrison (1949), in his discussion of a free-living species, 
Ctena,ytaina eucalypti, and its close association with lerp formers suggested that 
lerp-making habits may function as a way of surviving in sticky or gummy 
surroundings. Although C. eucalypti is not a lerp maker, it shares many peculiar 
104 
structures with true lerp makers. Heslop-Harrison believed that some of these 
peculiarities are undoubtedly closely associated with the lerp-making habits. This 
suggests a variant of the microenvironment hypothesis for evolution of lerps, 
contrasting to avoidance of desiccation, the focus of previous theories. The 
flocculence or lerp seems to provide a way of carpeting that may serve both 
functions. In fact, observation of the lerp forming Celtisaspis and Tetragonocephala 
found that the lerp actually encloses the nymph even on the under side. If the 
lerp is formed solely to avoid desiccation or even natural enemies, one would not 
expect the lerp to cover the underside of the psyllid body beneath the leaf. This 
feature in Pachypsyllini lerp-formers, not gummy eucalyptus feeders, may be 
explained as a primitive characteristic maintained from the Eucalyptus-feeding 
ancestors. More careful observations are necessary for testing the hypotheses. 
The basal groups of Spondyliaspidinae, i.e. Ctena,ytaina, Phellopsylla and 
Phyllolyma, are either free-living or lerp-forming. The free-living species produce 
copious amounts of flocculence, and tend to be cryptic. Ctenarytainini live among 
leaf buds, within rolled leaves caused by spiders or caterpillars, under deserted 
lerps of Spondyliaspidini, and even inside sawfly leaf mines (Taylor, 1990). 
Phellopsylla hide under bark and cover the space with fair amount of wax 
(Froggatt, 1990). Even the Ierp-formingPhyllolyma produce primitive lerps similar 
to some Glycaspis species although it is more like a sugary encrustation than the 
true lerps of Cardiaspina and Lasiopsylla (Morgan, 1984; Froggatt, 1990; Taylor, 
1990). This suggests that these taxa are intermediate between the free-living and 
105 
true lerp-forming groups. 
106 
1:1, 
, I:! 
t: i: 
t 
·I 
I 
I, 
,, 
.I 
I 'I 
I 
.,1 
--- . - ·-=--- ~ 
Appendices 
107 
Appendix 4-1. Specimens examined* for the Study of Pachypsyllini/ 
Spondyliaspidinae Phylogeny. 
Pachypsyllini Becker-Migdisova 
Pachypsylla Riley [see Table 1-3, USNM, MMY] 
Tetragonocephala Crawford [10 dry USNM, 20 d'+ 9 + n MMY] 
Celtisaspis Yang & Li [1 dry USNM, 15 d'+9+n MMY, 5 dry FSL] 
Spondyliaspidini Schwartz 
Cardiaspina Crawford 
C. albitexturata Taylor [2 slds, nymph+ d'+ 9, DH; l d' + 19 dry prtp, 
USNM] 
C. vittafonnis [l sld, d'+ 9, DB] 
Creiis Scott 
C. tecta [l sld, d', DB] 
C. longipennis (Walker) [2d'+29 lerps, KLT; d'+ 9 +n, 3 slds, ANIC] 
Glycaspis Taylor 
G. (Glycaspis) baileyi Moore [1 sld, nymphs+ d' + 9, DH; 3d' +3 9 
lerps, KLI] 
G. (Synglycaspis) aggregata Moore [1 sld, d', DH; 2 slds, d' + 9, DB] 
G. (Synglycaspis) planitecta [l sld, d', DB] 
Phellopsylla Taylor 
P. sp. [2d' + 2n, 4 slds (3plvp ), ANIC] 
P. sp. [ 49 + many n, in ale, GT] 
Phyllolyma Walker 
P. (=Cometopsylla) rufa (Frog.) [2d', 2 slds, ANIC] 
P. sp. [ d' + n, 2 slds (plvp ), ANIC] 
P. sp. [a+n, GT] 
Spondyliaspis Signore! 
S. plicatuloides (Froggatt) [ d' + 9 + n, 3 slds, ANIC] 
Ctenarytainini White & Hodkinson (1985) = Spondyliaspidinae 
Ctena,ytaina Ferris & Klyver 
Others 
C. eucalypti (Markell) [3 slds ( d', 9 ,n), DH; 1 sld, nymph, DB; 5 dry, 
USNM; 2d', 2 slds (plvp ), ANIC] 
Arepuna Tuthill = Russelliana (Aphalaroidinae) 
Acizzia Heslop-Harrison 
A. uncatoides (Ferris and Klyver 1932) [3 slds ( d', 9 ,n), USNM 
(Brown's collection)] 
Colophorina Capener (Euphalerinae sensu Hollis ms) 
C. cassiae Capener [1 dry ( d'), DH] 
Cometopsylla Froggatt = Phyllolyma (Spondyliaspidinae) 
108 
I Ii 
i, 
I I, 
; 11 
\\ 
'" ,1 
" 
.,1 
:1 
', 
Euphalerus Schwartz (Euphalerinae) 
E. sp. A(=? Colophorina) [3 slds (d,9,n), DH] 
E. nidifex Schwarz 1940 [ d+ 9 117 dry NE, 14 dry NT, 
lerps + nymphs USNMJ 
Heteropsylla Crawford 1914 
H texana Crawford 1914 [6+ slds, d+ 9n, USNM] 
Russelliana (syn. Arepuna Tuthill) = Aphalaroidinae 
R adesmiae [4 slds, d+ 9 prtp, DB] 
R fabianae [4 slds, d+ 9 prtp, DB] 
Trigonon Crawford 1920 
T. longicomis (Crawford 1919) [ 4 dry (3d,29), USNM Tuthill 
collection J 
* Acronyms: 1) prtp: paratypes, slds: slides, dry: pinned specimen; 2) ANIC: 
Australian National Insect Collection, DB: Daniel Burckhardt, Geneva Museum, 
DH: David Hollis, National Museum of Natural History, London, FSL: Fasheng 
Li, Beijing Agricultural University, China, KLT: Keith L. Taylor, CSIRO, MMY: 
Man-Miao Yang, USNM: U.S. National Muesum of Natural History; YJK: Yong 
Jung Kwon, South Korea; 3) Biogeographic regions: NE: Neartic, NT: Neatropic, 
ET: Ethiopian, OR: Oriental, PA: Paleartic, AU: Australian. 
109 
I' Iii 
,, ' 
I,'. 
!•1 
• I 
I 
:i 
ii 
1/ 
.,I 
:I 
'I 
Tables and Figures 
110 
I'. 
'' lie 
•1tl 
11 , 
:11 
: 
i1, 
T 
ii 
.,, 
,I 
II 
\\ 
.,I 
,..... 
,..... 
,..... 
Table 1-1. Species entities described in Pachypsylla (Homoptera: Psylloi<lea). 
SPECIFIC GALL POSITION HOST PLANT 
EPITHET AND SHAPE CELT!S*** 
celtidismamma* + leaf occidentalis 
(Riley 1876) mammiform (laevigata) 
(temaf olia) 
celtidispubescens leaf reticulata 
Riley 1890 mammiform laevigata 
celtidisglobulus + leaf sp. 
Riley 1890 mammiform 
celtidiscucurbita leaf sp. 
Riley 1890 mammiform 
celtidiscucurbita var? leaf sp. 
(Riley 1890) mammiform 
sp. leaf laevigata 
Riemann 1961 *H mammiform 
rohweri leaf reticulata 
Cockerell 1910 mammiform 
celtidisminute leaf ? 
Mally 1894 ? 
celtidisasteriscus + leaf laevigata 
Riley 1890 star-shaped reticulata 
tenuif olia 
(occidentalis) 
~-- = :_ ___ :. _;_= _._._-
~-~"~---=-::=..::..=.-:=. 
DISTRIBUTION OVER-
WINTER 
AZ CO CT IA IL ID IN KS adult 
MD MN NE NJ NY NC OH 
OK TX UT VA DC CANADA 
MS TX OK CO adult 
MSMO adult 
MOIA adult 
MOTX adult 
TX adult 
CO TX OK adult 
IA KS adult 
MD MS IA OK TX VA MO adult 
LA 
I-' 
I-' 
N 
Table 1-1 ( cont.). Species entities described in Pachypsylla (Homoptera: Psylloidea). 
SPECIFIC GALL POSITION HOST PLANT 
EPITHET AND SHAPE CELTIS*** 
c el t idisumbi Ii cus leaf sp. 
Riley 1890 wart-like 
depressed at middle 
celtidisvesicula* + leaf occidentalis 
Riley 1884 blister-like (reticulata) 
(laevigata) 
new sp. 1 leaf laevigata 
Riemann 1961 **+ blister-like 
new sp. 2 leaf reticulata 
Riemann 1961 **+ blister-like 
new sp. 3 leaf reticulata 
Riemann 1961 **+ blister-like laevigata 
new sp. 4 leaf reticulata 
Riemann 1961 **+ variously modified laevigata 
ass. w / others 
venusta * + petiole occident a/is 
Osten-Sacken 1861 spherical laevigata 
reticulata 
tenuifolia 
trident ala petiole sp. 
Patch 1912 
___ :.__ __ :..:.._= 
_ -_---=-=-=-=--== 
DISTRIBUTION OVER-
WINTER 
IA MS MD VA MO AR adult 
AZ CT IL IA KS LA MD MS adult 
NE NY OH OK VA 
TXOK adult 
TX NMAZOK adult 
TX adult 
TX adult 
CO CT IA ID KS MD MS NJ 5th instar 
NM NC OH TX TN VA nymph 
co ? 
f-' 
f-' 
w 
Table 1-1 ( cont.) . Species entities described in Pachypsylla (Hornoptera: Psylloidea). 
SPECIFIC GALL POSITION 
EPITHET AND SHAPE 
celtidisgrandis petiole 
Riley 1876 
celtidisgemma * + bud 
Riley 1885 irregularly round 
celtidisinteneris * + twig 
Mally 1894 obolong-oval 
ungulata twig 
Caldwell 1938 obolong-oval 
dubia* ? 
Patch 1912 
pallida* bud 
Patch 1912 irregularly round 
covered with dense hairs 
tropicala ? 
Caldwell 1944 
* Species recognized by Tuthill (1943) . 
+ Species recognized by Riemann (1961). 
*'' Unpublished Ph.D. dissertation. 
*'' '' Host plants in parantheses are probably erroneous records. 
HOST PLANT 
CELT/S*** 
? 
occidetalis 
laevigata 
retiwlata 
tenuif ofia 
occidentalis 
tenuif olia 
laevigata 
retiwfata 
sp. 
sp. 
reticulata 
? 
DISTRIBUTION OVER-
WINTER 
IA ? 
CT IA KS MD MO MS NY NJ 5th instar 
OH OKTX VA DC nymph 
MD IA IL OH KS TX VA 5th instar 
nymph 
OH 5th instar 
nymph 
TX ? 
AZNMTX ? 
MEXICO ? 
,..... 
,..... 
+>-
Table 1-2. Synopsis of hackberry psyll i<l gall types, Paclzypsylla (Homoptcra: Psylli<lae), their distribution and association with host species. 
GALL TYPES ASSOCIATED DISTRIBUTION ASSOCIATED 
SPECIES NAMES HOST PLANT* 
PETIOLE GALL P. venusta (Osten-Saken 1861) All over the hosts ranges, northeast to C. occide11talis 
P. tridentata Patch 1912 CT, southwest to AZ, UT, southeast to C. te11uif olia 
P. celtidisgrandis Riley 1876 Georgia, northwest to ID. C. /aevigata 
C. reticulata 
GLABROUS BUD GALL P. celtidisgemma Riley 1885 Central and eastern U.S. C. occide11talis 
C. te11uif olia 
C. /aevigata 
HAIRY BUD GALL P. pallida Patch 1912 AZ, NM, OK C. reticulata 
TWIG GALL P. celtidisintemeris Mally 1894 Reported from central and eastern U.S. C. occidentalis 
P. u11gulata Caldwell 1938 C. tenuif o/ia 
C. laevigata 
C. reticulata 
REGULAR BLISTER GALL P. celtidisvesicula Riley 1884 Northeastern U.S., as for C. occidentalis C. occidentalis 
without upper central spine 
new sp. 1 Riemann 1961 ** TX C. /aevigata 
new sp. 2 Riemann 1961 ** TX, NM, AZ, OK C. reticulata 
BLISTER GALL ? LA C. laevigata 
with upper central spine VA ( C. tenuif olia) 
ROUGH BLISTER GALL new sp. 3 Riemann 1961 ** TX C. reticulata 
without upper central spine (C. laevigata) 
DISC GALL P. celtidisumbilicus Riley 1890 MD, VA, AR, MO, IA, MS C. occidentalis 
..... 
..... 
V1 
T able 1-2. ( cont.) Synopsis of hackberry psyllid gall types, Pachypsylla (Homoptera: Psyllidae ), their distri bution and association with host species. 
GALL TYPES ASSOCIATED DISTRIBUTION ASSOCIATED 
SPECIES NAMES HOST PLANT* 
HAIRY NIPPLE GALL P. celtidismamma (Riley 1876) Widely distributed in eastern North C. occidcntalis 
America, north to Ontario, south to NC, 
east to CT, NY, west to NE, KS, OK. 
P. rohweri Cockerell 1910 CO, TX, OK C. reticulata 
P. celtidispubescens Riley 1890 Southwestern U. S., e.g. TX, OK, CO, C. reticulata 
MS. 
GLABROUS NIPPLE GALL P. celtidisglobulus Riley 1890 Northeastern U.S., e.g. MD, VA, MO, C. tenuif olia 
MS. 
P. celtidiscucurbita Riley 1890 
P. sp. Riemann 1961 ** (probably= Southern U.S., e.g. MO, TX. C. /aevigata 
P. celtidiscucurbita var. ? R iley 1890) 
STAR GALL P. ce/tidisasterisca Riley 1890 Southern U.S., e.g. TX, MO, LA. C. /aevigata 
with upper central spine C. reticulata 
STAR GALL P. ce/tidisasterisca Riley 1890 Northeastern U.S ., e.g. MD, VA. C. tenuif olia 
without upper central spine ( C. occidentalis) 
UNKNOWN P. tropicala Caldwell 1944 Mexico unknown 
P. dubia Patch 1912 TX unknown 
INQUILINE new sp. 4 Riemann 1961 ** in TX and may be widespread in all gall C. laevigata 
types except occide1Ztalis blister gall. C. retiwlata 
C. occide1Ztalis 
C. te1111if olia 
* host in parentheses = occasional observation. 
** based on unpubli shed dissertation of Riemann (1961). Proposed species here numbered arbitrarily. 
Table 1-3. 
Localities from which samples of populations of different Pachypsylla gall types 
from various host plants were obtained for protein electrophoretic, life history 
and morphological studies. 
LEAF ST AR GALL (LS) 
Localities sampled1 
MD: BB2.b, NAL2.t>, PG•, PRP•, SI" 
VA: WBP" 
GFV2,a 
TX: Aus· 
(C. tenuifolia). 
(C. laevigata); 
(C. occidentalis). 
(C. laevigata). 
LEAFG 
LABRous NIPPLE GALL (LNg) MD: BB2,b, DEAL·, NAL2.b, PRP', PG·, SI· 
LEAp l-IA.IRY NIPPLE GALL (LNh) 
LEAF DISC GALL (LD) 
LEAF BLISTER GALL (LB) 
116 
VA: GFV2.b 
WBP• 
AR: FAY" 
LA: ALX• 
MD: BB•, CMt2·b 
VA: GFV• 
OH: OARDC2·• 
AR: FAY2·• 
MD: GFV2.t> 
AR: FAY• 
(C. tenuifolia). 
(C. occidcntalis); 
(C. laevigata). 
(C. occidentalis) . 
(C. laeviga ta). 
(C. occidentalis). 
(C. occidentalis). 
(C. occidentalis) . 
(C. occidentalis). 
(C. occidentalis). 
(C. occidentalis). 
MD: CMt2·b (C. occidentalis). 
VA: GFV2·b (C. occidentalis) . 
OH: BA yi.t> (C. occidenlalis). 
AR: FA y• (C. occidentalis). 
LA: ALX• (C. Jaevigata). 
TX: AVS2.b, BFL2.c, PMT2,c, ZKR2,c 
(C. Jaevigata). 
BFL2·C, PMT2.c, ZKR2.c 
(C. reticulat a). 
AZ: TUC2.a (C. reticulata)_. 
ID· PCT (C. occidental1s). 
ND: FRGc (C. occidentalis). 
Table 1 
·3. ( cont.) 
Gall~ 
PETIOLE GALL (P) 
GLABRous BUD GALL (Bg) 
HAIRY BUD GALL (Bh) 
TWJ G GALL (T) 
Localities sampled 1 
MD: BB2.b, SI2.a 
Hus2 .. 
TX: AUS2·c 
AZ: TUc2·c 
(C. tenuifolia); 
(C. occidentalis). 
(C. laevigata). 
(C. reticulata) . 
MD: BBc, HUS2.a, NAL2·c, PG• 
VA: GFV2.b 
LA: ALXC 
OK: BSP2·c 
OK: BSP2·c 
(C. tenuifolia). 
(C. occidentalis). 
(C. laevigata). 
(C. Jaevigata?). 
(C. laevigata?) 
I A. 
cronyrns for 
/\.tb: B each population used: 
OJ-i: 
A.I<: 
lb: 
Nn. 
LA:. 
O.l<:; 
Tx: 
BB- Branchville, Berwyn; CMt- Catoctin Mountain Park, Thurmont; DEAL-
Li~~adwater R~., Churchon; HUS- Hughes Rd, Seneca; NAL- National Agricu!tura/ 
p ary, Beltsville; PG- Schoolhouse pond, Upper Marlboro; PRP- Patuxent River 
G ark, Upper Marlboro· SI- Smithsonian Environmental Research Center, Edgewater. 
B_FV. Great Falls, McLean- WBP- George Washington's Birthplace, Washington 1rthplace. ' 
DBA y_ South Bass Is. Put-in-Bay OARDC- Ohio Agricultural Research and eve Jo ' ' F Pment Center, Wooster. 
p A y_ FayetteviIIe 
F CT. Pocatello. · 
RG. Fargo. 
_is\X- C~J_e Street, Alexandria. 
A' · Boi11~g Spring State Park, Woodward. . .. 
_Us. Austm city; BFL- Brackenridge Field Lab of Univ. of Texas at Austin, ZKR-
Zilker p k . nza1 TU ar , Austm; PMT- Palmetto Park, Go es. 
2 P C- Santa Rita Mts, Tucson. 
0Pulation 
a s sampled for morphological study. 
Only 
b l3 aduJt sam 
c 
0
t11 aduJt pies used for electropl10retic study. 
0 l and n h · d 
n Y nyrn h Ymp al samples used for electrophoret1c stu Y· 
P al sam 1 · d P es used for electrophoret1c stu y. 
117 
....... 
....... 
co 
Table 1-4. Enzyme loci scored for analysis of Pachypsylla. 
ENZYME E. C. # LOCUS 
F ructose-1,6-Diphosphatase 3.1.3.11 FDP 
Fumarate Hydratase 4.2.1.2 FUM 
G lyceraldehyde-3-Phosphate 1.2.1.12 G3PDH 
Dehydrogenase 
Glycerol-3-Phosphate 1.1.99.5 GPDH 
Dehydrogenase 
Isocitrate Dehydrogenase 1.1.1.42 IDH-1 
IDH-2 
Lactate Dehydrogenase 1.1.1.27 LDH 
Malate Dehydrogenase 1.1.1.37 MDH-1 
MDH-2 
Malic enzyme 1.1.1.40 ME 
Peptidase 3.4.11 PEP-1 
Phosphoglucomutase 2.7.5.1 PGM 
6-phosphogluconate Dehydrogenase 1.1.1.44 6PGDH 
Phosphoglucose Isomerase 5.3.1.9 PGI 
Triose Phosphate Isomerase 5.3.1.1 TPI 
BUFFER VOLTAGE RUN TIME 
Tris-Citrate O.lM pH 8.2 180 1 hr 
Tris-glycine 0.025M pH 8.5 180 1 hr 
Tris-Citrate O.lM pH 8.2 180 1.5 hr 
Phosphate 0.02M pH 7.0 180 1 hr 
Tris-Citrate O.lM pH 8.2 180 1.5 hr 
Tris-Maleate-EDTA-MgCl2 0.05M pH 7.8 180 35 min 
Tris-Maleate 0.015M pH 7.2 180 1.5 hr 
Tris-Maleate-EDTA-MgCl2 0.05M pH 7.8 180 35 min 
Tris-Maleate 0.05M pH 7.8 180 1.5 hr 
Tris-Glycine 0.025M pH 8.5 180 45 min 
Citrate Phosphate O.OlM pH 6.4 180 1.5 hr 
Tris-Maleate-EDTA-MgCl2 0.05M pH 7.8 180 1.5 hr 
Citrate Phosphate 0.01M pH 6.4 180 1hr 10 min 
Tris-Citrate O.lM pH 8.2 180 1 hr 
N ;"~ 0- .... -- ..... 0 .... 8:;:!t;;:;:!g 
. . . . . . . .,~~~~~~~ .,g ggg .,ggggg "'88~~8~~ .,~~~ ., ~~8~ .,~~ 
~~~ -og;g~g;g .,gg~~8§8 .,~~~~ 00000 .,~~~~~~8 «>~~~ ., ~~&~ .,~~ CIOOOOOO . . . . . . . . ..... . q~qqq 
w 
Q< .,~~~~~~~ co§§~8§§§ o ..... c,o 00000 0000000 000 .0 8888 .,~~ 0- X W '°88~8 10~~~~~ cog8ggggg CIOO OO -~o .... .. . 
. . . . . . . 
000 ~~~~ 
«>x- «>§!§~§§§ t0§§~~§§§ '°~~~ ~ co~~~~~ co ~~~~~~~ io§ §§ .,~~~~ co~ ~ ....... . .... . . 
~~- «> §8§~§ !9 § .,gg::c~~gg "'"'"'"'"'0 cog§§§§ «> §§§§§§§ 000 .,~~~~ "'~ 8 CI N.,0 .- 0 co~ ~~ ....... ~~~-:--=~~ - o«>o . . . ... . 
Q- ., ~ ~~~ ~~~ oo.:i,,..,ooo ., ~~~~ .,gg~8§ ., ~HJ~§8§g ., ~~~ "' ~8~~ -o~§ -ox~ ., ~~~~~~~ ~ ... . . .. 
~ '° ~~~ ~~~~ 
- -
'°~~~~~~~ '°~~~~ '°~ 88 "'~~~~ ~~~ -o§g~g§ -oggg§ggg ....... 
~ Q~ 
! ID ~~~~~ ~~ .,~~~~~~~ 0 "'"'Cl) 0 .,ggggg co §§§§§§§ co§§§ '°~~~~ "'88 ~~~ cog 8~8 ~~~~~ . .... . . § 
. ~ .,~~~~~~~ .,~~~~~~~ .,~ ~~~ «>§§ ~§~ .,ggggggg '°~~~ «>§§§§ .,~~ ~ "'~~ 
!: ! ....... 
It ! 
x< ~~E~§~g~g ~~~~~~~~ ~ ~~~~ ~ggggg ~ ~~ ~ ~~~~ ~§§§ ~8888 ~~ ~ N~ X . . . . . . . 
~< .nggggggg "'~~~~~~~ ~ §§~~ 00000 "' ~~~~~~~ 000 ~~~~~ "' ~~ .n ooooo .n ooo -~x ... . ... ~~ ~~~ 000 o,e 
H Q- ~~~~~~~~ ~88~8§8§ ~ ~~~~ ~~~~~~ ~ ~~~~8~8 ~ ~g~ ~ ~~~~ ~ ~~ o~m ....... 
H M• .,~~~~~~~ .,~~~~~~~ ., ~~~~ "'~8~8~ ., ~~~~~~~ ., ~~~ "' 88~~ .,~~ ll!"2 .. ~-g • 
.:: i ~ - "'~ ~~~8~~ "'~ ~~~~~~ P"l gggg ,.,~~~~~ "' ~~~8~88 "'~ 8~ "' 8~~8 "' ~~ i; ., ~ -
~::. 
.... ~~ ;:~ ~~~§~~ ~~~ ~~8~~ ~ ~~~~ ;: 8888~ ~~~8~~~~ ~ ~~g ~ ~~~~ ~88 - . :-2 
H -o~g o-§~ § ~~~~ .. ~~~~§~~ o- §§8§ °' ~~~~ ~ o. §gggggg .. ~~~ .. ~~~~ .. ~~ . . . . . . . . . . .... 
,;;< 
~i Q> .. ~ ~ ~~ ~~~ .. ~~~~~~~ .. ~~~~ ., ~~~~~ .,ggggggg .. ~~~ -o~§~~ x-_.., ...... . 
!S 
!I ~~~ ~~::&$8~~ ~~~~§~~~ ~8888 OOP"IO,.._ ~gg~g~gg ~ ~~~ "'~888 . . . . . . . ~ ~~~~~ ....... fi ~g~~:§§§8 ~§~E;~§§§ ~ ~~~~ ~~~~~~ 0000000 ~~~~ .,~~~~ ~ ggggggg . . . . . . . . .... . . .... . .. i~ 
> ~ ~~~~~~~ ~~~~~~~~ ~gggg Nggggg 0000000 ~~~~ ~88~~ .!: : N~ U 00000000 "')~~~~~ N OOOOOOO i~ . ...... 
<" ~8§§~§~§ 2:§§S:8~8§ 2: ~~~~ ..,ggggg 0000000 e ~~~ .... ~~~~ 2:§ NOOO O OOO N~~~~~ .- 0000000 . . . . . . . . . . . . . . ... ... . 
~ 
i 3E<muow~uu~<muow ... uf8<muow8<muow38<muow-u~S<~u~€<muo~S<m 0. a.. u X ... .._ - u 
119 
.,~~~~~~ ~ ., ~~~~~gg 
., ~ ~~~~~ ~ 
... ~~~~~~~ 
<>:11§:Hl§'.H! '°~~~ ~~~~ . . .. . . . 
.. ~~~~~g~ 
.. ~~~~~~8 
.. ~ ~~ ~~~~ ~~~~ ~~~~ 
.,ggggggg 
.. ' ... . 
'ooooooo 
~ggggggg 
. .. .... 
00 .......... 000 
~880:~ggg 
.. ..... 
OC>C>O 
o.gggg 
OC>OC> 
aooooo 
0011'\II'\ 
.. ~~~ ~ 
"'~ 8~8 
'°~ ~~~ 
ll) gggg 
~~ ~~~ 
0000 
tOOC>OO 
0000 
120 
00000 
~ggggg 
o,~gg~~ o.§§§§§8§ ~ggg o.gggg '°g§ 
'' ' .. -~ . . ... -~ . . . . ·~ ·~ 
ll)ggggg ~ggg~~~g -a§§§ ~gggg "' gg 
~ ...... .. . .. -~ .. -~ . -~ 
~ggg ~gggg 
. . . . . .. 
.. ~~&~~ "' &~~g~gg .. ~~~ ., ~~~~ ., ~~ 
"' ~8~~~ "'8~~8&~8 "'8&~ .. ~~~~ 
'°~~~~~ .., ~~~~~8~ <> ~gg ..,~~~~ <> g~ 
tO~~~~~ tO~~~~~~~ io§§§ 
co§§§§§ co §§;;§§~§ 
.,~~g ~ ~~~~ ....... 
Tablt 1·5 (cont.) 
·· · ··-····· ··· · ··- --·-····-··-·-· · ·--· ---- ·· -···----------- --·-- ------- --- -----·-----------·--·-
Popul1tton• 
-- --·---··--··- ·-· ·· ·- ··-----·- ··--- ·- ·· ·-· ··-·-· ·---· ----- -- ·--- ·--·-·---··--··· --··-- ·-··-· ·--· ··················- -- ---····· 
4 5 6 7 8 9 10 11 12 1J 14 15 16 17 18 19 20 21 
LMh LS LD LB , .. LNh LB LNh LS LN• LS LM• LS 
"' 
LS 
"• 
lS LNg LMg LS LMg 
Locus (CfV) (GfV) (G FV) (CfV) (GfV) CCM t) CCJH) (88) (88) CBB> (MAL) {MAL) (PRP) (PRP) (PC) (PG} (SI) (SI) (OEALE) (\IBP} (WP) 
--·---·-· ····· ······ ············---------·---·------·······-----·----- -----·-·-· ···· ·-----·-·--- ·-··-···· · ···· ...... ·····- ····--- · · · ---··· ·- ·· 
IOH·t 
(Ml 21 29 22 22 8 6 ,, 3 8 12 5 ,, 8 8 6 8 8 8 8 8 8 
A .000 .000 . 000 .000 . 000 .000 .000 .000 . 000 .000 . 000 .coo . 000 .000 .000 .000 .000 .000 .000 .000 .000 
8 . 190 .362 .023 . 068 .063 . 000 .000 .000 . 125 .000 .200 .000 .000 .000 .250 .000 . 188 .000 .063 . 188 . 000 
C .500 .224 .159 .864 .250 .250 .727 . t67 . 313 .042 .000 .000 .250 .125 .250 . 000 .J1J . 063 .063 .063 . 000 
0 .000 . 000 .000 .000 . 000 . 000 .000 .167 .000 .083 . 200 .000 .000 .000 . 000 .125 .000 .063 .000 .000 .000 
E .310 . ,1, .795 .045 .625 .667 .273 .667 .500 .873 .600 1. 000 .688 .875 .500 .m .500 .875 .875 .730 1.000 
' 
. 000 .000 .000 .000 .000 .000 .000 . 000 .000 . 000 . 000 .000 . 000 .000 .000 .000 .000 .000 .000 . 000 .000 
' 
.000 .000 .023 .023 .063 . OBJ .000 .000 .063 .000 . 000 .000 .063 .000 .000 .000 .ODO .000 .000 .000 .000 
IO H-2 
(M) 23 31 24 19 10 6 ,, 3 8 12 5 12 8 8 6 8 8 8 8 8 8 
A .000 .000 .000 .000 .000 .000 . 000 .000 .000 .coo .000 .000 .000 .000 .000 .000 .000 .000 .000 . 000 .000 
' 
.000 .000 .000 .000 .000 .000 .000 .000 .000 .ODO .000 .0,2 .000 .000 .000 .000 .000 .000 .000 .000 .000 
C .000 .016 .000 .000 . 000 .000 .000 .000 .063 .04Z . 000 .000 . 125 .DOO .167 .06J .125 .000 .063 . 000 .000 
D 1.000 .984 1.000 1. 000 1. 000 .917 1.000 .833 .938 .958 1.000 . 917 .875 1.000 .833 .875 .873 1.000 .938 1.000 1.000 
E .000 .ODO . 000 . 000 .000 . 000 .000 .000 .000 .000 .000 .042 .000 .ODO .000 .063 . 000 .000 .000 .000 .000 
' 
. 000 .000 .000 .ODO .000 .000 .000 .000 . 000 .000 .000 .000 .000 .coo .000 .000 .000 .000 .000 .000 . 000 
' 
.000 . 000 .000 .ODO .000 .083 .000 . 167 . 000 .000 .000 .000 ,000 .000 .000 .000 .000 .000 . 000 .000 . 000 
H .000 .000 . 000 .ODO .000 .000 . 000 .000 . 000 .000 . 000 . 000 .000 .000 .000 .000 .000 .000 .000 .000 .000 
)o'!)H-1 
(M) 5 13 3 2 3 7 7 3 7 ,, 5 12 2 2 6 8 8 8 8 8 8 
A .000 .000 ,000 .000 .000 . 000 .071 .000 .000 . 091 .000 ,042 .000 .250 .000 . 188 .000 .000 . 188 .000 .000 
I--' B . 000 .038 .167 .500 .000 .214 .429 . 000 .214 .045 . 100 .208 .000 .250 .000 .1 25 .063 .063 .063 .000 .125 N C . 000 .000 . 000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .t2S .000 .000 .125 .000 .000 
I--' D 1. 000 .923 .8JJ .500 1.000 .786 .500 1.000 .786 .818 .900 .667 1.000 .500 1.000 .438 .938 .938 .625 1.000 .87'5 
.000 .000 .000 .000 .ODO .000 .000 . 000 .000 .000 .000 ,000 .000 .000 .000 .125 .000 .000 .000 . 000 . 000 
.000 ,036 .000 .000 .000 .000 . 000 .000 .000 . 0<5 .000 .OBJ .000 . 000 .000 .000 .000 .000 .000 .000 .000 
.000 . 000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 . 000 .000 .000 
.000 . 000 .000 .000 .000 .000 .000 .000 .000 .000 .000 . 000 .000 .000 .000 .000 .000 . 000 .000 . 000 .000 
.000 .000 .ODO .000 . 000 .000 .000 . 000 .000 .000 . 000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 
KlK·2 
(M) 7 12 8 7 6 9 12 3 8 12 5 12 6 8 5 8 8 8 8 6 8 
A .000 .125 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 
1 . 000 . 875 1.000 1.000 1.000 1.000 1.000 1 . 000 1.000 1.000 1. 000 .917 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1. 000 1.000 
.000 .000 .000 .000 .000 . 000 .000 .000 .000 .000 .000 .083 .000 . 000 .000 .000 . 000 .000 . 000 .000 .000 
.000 .000 .000 .000 .000 .000 .000 .000 .coo . 000 .000 . 000 .000 .000 .000 .000 .000 .000 .000 .000 .000 
.000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .ODO .000 .000 .000 .000 . 000 .000 .000 .000 .000 .000 
.000 .000 .000 .000 .000 .000 .000 .000 . 000 .000 .000 . 000 .000 .000 .000 .000 .000 .000 .000 . 000 . 000 
. 000 .000 .000 .000 .000 .000 .000 .ODO .000 .000 .000 .ODO .000 .000 . 000 .000 .000 .000 .000 .000 .000 
H .000 . 000 .000 .000 . 000 .000 .000 .000 .000 .000 . 000 . 000 .000 .000 .000 .000 .000 .000 .000 .000 .000 
I .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .ODO .ODO .000 .000 .000 .000 .000 .000 . 000 . 000 
6PGOH 
"' 
10 18 
" 
9 6 8 5 2 8 9 5 9 8 7 6 8 8 8 8 8 8 
.000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 
.050 .028 .000 .000 .000 .000 .000 . 000 .000 .000 . 000 .000 .ODO .000 .000 .000 .000 .000 .000 .000 . 000 
.050 .000 .000 .000 .000 .000 .000 .000 .000 .ODO ,000 .000 .000 .000 .000 .000 . 000 .000 .000 .000 .000 
.850 .861 .8'57 .611 1.000 .938 . 500 1 .000 1.000 1.000 1. 000 1.000 .9)8 1.000 .BJ) .938 .938 1.000 1.000 1.000 .938 
.000 .000 .000 .000 . 000 .000 .200 .ODO .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 
.000 .056 .0 71 . 222 .000 .000 .,oo . 000 .000 .000 .000 .000 .063 .000 . 167 .000 .000 .000 .000 _ooo .063 
.050 .056 . 036 .167 .000 .06) . 000 .000 .000 .000 . 000 .000 .000 . 000 .000 . 063 .063 . 000 . 000 .000 .000 
.000 . 000 .036 .000 .000 .000 .000 . 000 . 000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 
LOH 
(M) 2) 
" 
25 24 8 9 12 3 8 12 5 12 8 8 6 8 8 8 8 8 8 
.000 .000 .ODO .000 .ODO .000 .000 .000 .000 . 000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 
,000 .000 .000 .000 .000 .056 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .ODO . 000 
.261 .371 .O<O . 188 . 188 .056 .042 .000 .063 .000 .000 .000 .000 .125 .000 .063 .000 . 000 .063 . 125 .000 
.500 .500 .880 .604 . 730 .m .792 . BJJ .730 .873 .900 . 958 1,000 .875 .917 .938 .730 .938 .873 . 813 .688 
.,1, .081 .060 .146 .063 .000 .083 .167 .000 .042 . 000 .000 .000 . 000 .000 .000 .000 .063 .000 .063 . 063 
.065 ,Oli8 .020 . 042 . 000 .1 11 .OBJ .000 . 188 .083 . 100 .042 . 000 .000 .000 .000 .250 .000 .063 .000 . 250 
.000 .000 .000 .021 .000 . 000 .000 .000 .000 .000 .coo .000 .000 .000 .083 .000 .000 .000 . 000 .000 . 000 
-··· · ·····-···········-·············· · ·· · -· · ···-·······-····· · ··· · ···-··-····- · ··· · · · ······-------------------· --- · ·· ···· · ···- ·····-···--· ·· · ·-··---- --- --· · ·- · 
• ue uble 3 for 1cronyms for gal t types ind popul1tlons. 
~- '°~~~~~~~ 00000000 '°~~~~~~~~~ -oggggggggg -oggg§gggg ~~~8~~~~ 0- W X ,o~~~~~~~~ ~~ u .. . .. . ... . .... . . . 
~ oaG' 0000000 00000000 N~~~~~~~~~ N~~~~~~~~~ 00000000 N~~~~~~~ m~ N g~g~~88 N~~~~~~~~ N~~~~~~~~ . .. . ... 
> 
-.#g~~~~gg ~gggggggg 
-88§§ggggg -~~~~~~~~~ -§§§ggggg ~~~~~~~~ "~-~m~ . . . . . '. -~~~~~...;~ . . . . .... . . ' .. .. . . 
~a.; N~~~~~~~ N~~~~~~~~ Nggggggggg ... ggggggggg N ~~~~~~~~ N~~ ~~~~~ . . ... . .. . . ' . .. . . . . 
-§§~~§§§ 00000000 
-§§§~g~ggg 000000000 00 ..... /""10000 
- §§§§§§§ ~~ ~ - 00000000 -4'000000000 ,..,~~~~~~~~ . . . . . . . ~~~~~~~~ . . . . . . . . . ~~~~~~~~~ ... ' .. . 
~~~ 0-~~~~~~~ 0000000-- 0-~~~~~~~~~ 000000000 0-~~~~~~~~ (>~~~~~~~ °'~~~~~~~~ "'1000000000 ~~~~~~~~~ 
0000000 00000000 co§§§i§~§§§ 000000000 Oa::101""10000 N~~ggggg ~a.~ cooooooo o co~~~~~ ~~~ "~~~~~~~~~ coo ..... 0..00000 0000000 ~°:~~~~~~ . . . . . . . . . . . . . . . . . . ... .. 
~- 0000000 oo ,..,cooooo 
..... ggggggggg 000000000 o ..... ocooooo 000 ............... _ NX~ aogg ggggg IO~~~~~~~~ a:>~~~~~~~~~ co~~~:i:~~~~ "888:-A;:!;:!~ M~ ~ . . . . . . . . . . . . . . . . ....... 
0000000 IO gggggggg 
..... g~g~ggggg 000000000 ~~~~~~~~~ ~ggg~~gg co ooooooo coggggggggg 000 0 000 C!C!C!~~~~~ - . . . . . . . . . . .... .. . ... . .... ... . ... 
·s ~ "°~~~~~~~ "°~~~~~~~~ .,..,~~~~~~~~~ '°~~~~~~~~~ "°~~~~~~~~ .,~~8~8~~ omm "'"'-' 2 } u 
.,..,~~~~~~~~~ .,~~8~~~~ :t3i 1()88~8!§§ aogggggggg -o;~ggggggg ao§§§~§ ~§§ . . . . . . . ....... . . . . . . . . . . . .. ..... 
Q- 0000000 
..agggggggg ..agg~~~~ggg '°~~~~~~~~~ '°~~~~~~~~ -oggg~;g~ .,xw -oggggggg N -' ~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . 
"-•w I()~ ~~~~~~ IO~~~~~~~~ ~§~:!i~~ggg co~~~~~~~~~ .,~~~~8~~~ "°~~~~~~~ ...... _,~ . . . . . . . . . 
~ow IO~~~~~~~ co§§§§§§§§ co~~~~~~~~~ ooggggggggg co §§§~§;8:C! co~~~~~~~ N _, ~ .... ' ... . . . . . . . . . . . . . . . . . 
~- 00 ............... 00 00000000 co§~§~§§§§§ co§§§§§§§§§ co§§§£;§~§§ co~~~~~~~ ~xw co~~~~~~~ co~~~~~~~~ N_,~ . . . . . . . . . . . . . . . . . . . . . ..... 
~mN oo,.,,o~oo 00000000 OOOIOONOOO 
... g~~gggggg ~~~~~~~~~ ~ggg~~~g N~~ ~~~~~°:~~ ~~~~~~~~~ ~g ggtg~ggg -c:"':~~~0:C!O:O: . . . . . . . . . .... .. . 
~~K 0000000 00000000 000000000 000000000 ~~~~~~~~~ ~~~~~~~~ NOQOOOOO N 00000000 N ooooooooo NOo ooooooo N~- - o:~o:o:o:o:~ -o:~o:c:o:~~~ -000000000 .-000000000 .. . ... . .. . ...... . . 
N~K O OO OIIJON N gggggggg ooor,...0 ,..,000 ~~~~~~~~~~ ooo~o ..... oo 000~ ........ -o NO OOO\AO..r ~g gg o:g:gggg NOOo- o cooo ~~~~~~~~ N~- -o:o:o:o:°:~~ - o:o:o:~c:~~o: ....... . . -o: o:c:o:~~~o: 
~ ... N ... N ::C 
~ ~~•muow~~~~cmuow~u::c~~•muow-ux-~§cmuow~u::c-~§cmuow-u::c~~cmuow~u 
122 
:: i 1 
-e 
i-1~ ! 
E 
0 50 
-= _§ 
f= 
.: ~ 
H 
H ~~ 
" ~ 8 -;: 
ri~ C ~ c -
.. 
~-
tl 
~gggggggggggg 2888§8§8 288§§§§88888 
.... · ..: ...... . · ..: . ... . ·.,.; ... . ... . ~g~g§~~~~ ~§~~~~~~~~ 
~~~~~~~~~ ~§~§~~~~~~ 
~gggggggggggg ~sgggggg ~g~gg~g~~~~; ~gg~g§~&s ~228§8§8§§ 
. . ... . .. ·.,.; . . . .. '. ·.,.; . . . . . . . . . . . . . . . . . . . . .. ·..: . . .. 
~gggggggggggg ~ggggggg ~gggggg;g~~~ ~gggggggg ~88888§8§8 
..... . .. ·.,.;.. . .... ·..: . . . . . . . . . . . '·..: . .. . . . . . ·..: ... . 
~gggggggggggg ~ggggggggg~~g ~~~~~~~~ 
' . . . . . . . . . . . ' . . . . . . . . . ' . -
123 
Tablt 1·6 . (cont . ) 
·· ······ ····· ·········· ·········· ···· ·· ··· -·· ·· ···-· ······ ··· · ·· ·· ·-· · ·· -······ ··· ·· ··· · · ··· ·· 
Popul et I oo .. 
························ ···· ·············· · ·· ··· ·· ········-······-· · ······ ·· ············· 
Locus 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Locus 16 17 18 19 20 21 22 
Call type LMh (i:t+c) Uth(sd) LWgCm+c) LJlg(cn) Ulg(sd) LS(m} LIC•) LBC•) LBC•> LB Cm) LB{m) LO ( m) p p p G1l ltypc 89 89 89 Bh T LRP LRP 
Local fty CHr CH? JI.U•SB GfV JtAL•BB NAUB8 OH CfV OH•JID• ID TX TX GFV TX BB AZ Loul ity LA MAL•B8 VA OK OK AZ CHI 
Host sp. C.o. C.o. C.t. C.o . C.t . C.t. C.o. C.o. C.o. C.l. C.r. C.o. C.l. C. t. c.r. Hos t sp. C, I. C.t. C.o . Cl? Cl? Cr c, 
· · --····--·· · ··· -· -··-·-· -···--··· · ·-···--·-- · --· · ···· · ··- · -· · · · · ·-· · ····- · ···· · · · ··· · · ················· · ······ · ·- · · ·····----· 
.... . .. 
FOP FOP 
(N) 2J 18 27 
' 
16 20 8 10 3 46 45 29 9 9 5 ( N) 6 5 J 12 8 5 10 
A .000 .000 . 000 .000 .000 .000 .000 .000 . 000 .000 . 000 . 000 . 000 ,000 .000 A , 000 .000 .000 .000 . 000 .000 1.000 
8 1.000 1.000 1.000 1.000 1. 000 1.000 1.000 1.000 1. 000 1. 000 1.000 1.000 1.000 1.000 1.000 B t .000 1. 000 1.000 1.000 1.000 1. 000 .000 
TPI !Pl 
(N) 2J 18 27 16 16 16 12 20 e 40 39 37 9 9 5 (N) 6 5 4 13 9 10 10 
A .000 . 000 .000 . 000 . 000 .OJI .000 .000 .000 .000 .000 . 000 .000 .000 .000 A .000 .000 . 000 .000 . 000 . 000 .000 
a 1. 000 1. 000 1.000 1.000 .969 .969 1.000 1. 000 1.000 .988 1. 000 1. 000 . 000 . 000 .000 a .000 .000 .500 . 000 .000 1.000 1.000 
C . 000 .000 . 000 .000 .ODO .000 .000 .000 .coo .000 .000 .000 .000 .000 .000 C .000 .000 .000 .000 1.000 .000 .000 
D . 000 . 000 .000 .000 . 000 . 000 .000 .000 .000 .013 .000 .000 1. 000 1.000 1.000 D .ODO .000 .000 .000 .000 .000 . 000 
E .000 . 000 . 000 . 000 . 031 . 000 . 000 .000 .000 .000 .000 .000 .000 . 000 .000 E 1.000 1.000 . 500 1. 000 .000 .000 .000 
C3POH GJPOH 
(M) 23 18 27 
' 
16 22 10 12 6 34 32 29 9 9 5 ( N) 5 4 4 e ! 10 10 
A .000 .000 .000 .000 .000 .000 .000 .000 . 000 .000 .000 .000 .000 . 000 .000 A .000 . 000 .000 .000 .000 .000 1.000 
8 t . 000 1 . 000 1.000 1.000 1.000 1.000 1.000 1.000 1 . 000 t .000 t . 000 1.000 1.000 1.000 1 . 000 a t.000 1.000 1.000 1. 000 1 .000 1.000 . 000 
IDH-1 IOH-1 
(M) JO 1! 39 16 22 27 18 22 14 42 45 41 9 9 5 (M) 6 5 4 12 9 10 e 
A .000 . 000 .000 .000 .000 .000 . 000 .000 . 000 . 000 .000 .000 . 000 . 000 . 000 A . 000 .000 . 000 .000 .000 .000 , Jr.; 
a .000 .000 .000 . 000 .000 . 000 .000 .000 .000 . 000 .011 .000 .000 .000 .000 a .000 .000 .000 . 000 .000 . 000 .625 
C .000 .000 .000 .000 . 000 .204 . 000 .000 .000 .000 .044 . 000 .000 . 000 .000 C . 000 .000 .000 . 000 .m .000 .000 
D .233 .389 .103 .000 . 023 .204 .9n .932 .71!6 1 .000 . 922 .134 1 .000 1 . 000 1. 000 D .000 .000 .000 . 000 .000 .000 .000 
E . 767 . 611 .872 .er.. .932 .500 . 02! .068 .000 .000 . 000 . 756 .000 .000 .000 E 1. 000 1.000 1 .000 1. 000 .000 .000 .000 
F .000 .000 .026 . 125 .000 . 093 . 000 .000 . 214 .000 .022 . 110 .000 .000 .000 
' 
.000 .000 .000 .000 .222 1.000 .000 
G .000 .000 . 000 .000 .045 .000 .000 .000 .000 . 000 . 000 . 000 .000 . 000 .000 G .000 .000 .000 . 000 . 000 . 000 .000 
\OM·2 IDH-2 
(M) 29 21 39 12 22 27 15 5 14 42 42 39 3 7 2 
"' 
5 4 2 5 7 10 10 
A .000 .000 .000 .000 .000 .000 . 000 .000 .000 .000 .000 .000 .000 .000 .000 A .000 .000 .000 .000 .000 .000 1.000 
8 .000 .000 . 000 .000 .000 .000 . 000 . 000 .000 .000 . 000 . 000 . 000 . 000 . 000 • .000 .000 .000 .000 . 000 1 .000 .000 C ,000 .000 .026 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 . 000 . 000 C .000 .000 . 000 .000 .000 .000 . 000 
0 1 .000 1.000 .9t..9 \.000 1.000 \. 000 1.000 1.000 1.000 1. 000 ,988 1.000 .000 .000 . 000 D .000 . 000 .000 . 000 . 000 .000 .000 
E .000 .000 .026 .000 .000 .000 ,000 . 000 .000 .000 .000 .000 . 000 .000 .000 E .000 .000 .000 . 000 . 000 .000 .000 
,...... 
' 
. 000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 . 000 .000 . 000 f .000 .000 .000 . 000 .000 .000 .000 
G . 000 . 000 .000 .000 .000 .000 .000 .000 .000 . 000 . 012 . 000 1 .000 1. 000 1.000 G 1.000 1.000 1. 000 1.000 1.000 .000 .000 N l'!lH·J• HOH · 1* 
+:>, ( N) 2! 21 33 16 19 16 13 10 8 38 37 37 9 9 5 ( N) 6 4 4 11 7 6 6 
A .000 .000 .061 .000 .000 .063 .000 .000 .000 . 000 .014 .000 .000 .000 .000 A . 000 .000 .000 .000 .000 .000 . 000 
• .071 .OZ4 .152 . 188 . 158 .219 .346 .100 . 063 .000 .068 , 054 ,000 .000 .000 • .167 .250 . 125 .045 . 000 .000 .000 C .036 .04! .000 . 000 .000 . 000 .000 .000 . 000 .000 . 014 .000 . 000 . 000 .000 C .000 .000 .000 .000 .000 .000 . 000 
D .8r.i . 405 . 742 .813 .632 . 719 .654 . 900 . r.;o .934 .824 .89Z .333 . 222 .400 D .833 . r..o . !75 .955 . 143 ,000 . 000 
' 
.000 .119 .000 .000 .079 . 000 . 000 .000 .000 . 000 . 000 . 000 . 000 .000 .000 
' 
. 000 .000 .000 .000 .000 .000 .000 
f .018 .381 .030 .000 .079 .000 . 000 . 000 .188 .066 .081 . 054 . 000 . 000 . 000 
' 
.000 . 000 .000 . 000 .000 .000 .000 
G .000 . 02, .015 . 000 .053 .000 . 000 .000 .000 .000 . 000 . 000 . 000 . 000 . 000 G . 000 . 000 .000 .000 .000 1.000 1.000 
H .000 . 000 .000 . 000 .000 .000 . 000 . 000 .000 .000 .000 .000 . 000 .000 .000 H .000 .000 .000 .000 .857 .000 . 000 
I .coo .000 .000 .000 .000 .000 . 000 .000 .000 .000 .000 .000 .667 .778 .600 I .000 . 000 .000 .000 . 000 .000 .000 
HOH·Z HDH -2 
(N) 26 18 39 14 22 27 14 20 14 46 45 39 8 9 5 ( N) 6 5 4 12 8 8 ! 
A .000 .000 .000 .000 . 000 .000 .000 . 000 .000 .000 . 000 . 000 .000 .000 .000 A .000 .000 . 000 . 000 . 000 1.000 1. 000 
a . 000 .000 .000 .000 .000 .000 .000 . 000 .000 . 000 . 000 .000 . 000 . 000 . 000 8 . 000 .000 . 000 .000 .000 .000 .000 
C 1.000 t .ODO 1.000 1. 000 1.000 1. 000 1.000 1.000 1. 000 1.000 1 .000 1.000 1.000 t.000 1.000 C 1. 000 1.000 1. 000 1.000 1.000 .000 .000 
6PCOH 6PCiOH 
(N) 28 21 39 12 22 27 16 1! 14 46 45 36 9 9 5 ( N) 6 5 4 13 9 10 10 
A .000 . 024 .000 .000 .000 .000 . 000 .000 .000 .000 . 000 . 000 . 000 .111 . 000 A .000 .000 ,000 . 000 .000 .600 .000 
a . 0 18 .024 . 000 . 000 .000 .000 .000 .000 .071 . 043 .011 . 000 .889 . 722 1 .000 • .000 .000 .000 .000 .000 . 350 .000 C .018 .000 .000 . 000 .000 .000 .000 .000 .036 . 01 1 .000 . 000 .ODO .000 .000 C .000 .000 . 000 .000 1.000 .050 .000 
·O .839 .952 . 987 1 .000 .!41 . 870 .813 .444 . 536 . 924 .933 . 76' .111 .167 .000 D . 000 . 000 .000 .000 .000 .000 .000 
E . 000 . 000 .000 .000 . 068 .019 .OJ\ .278 .000 . 000 .000 . 000 .coo .000 .000 E .000 ,000 .000 . 000 .000 . 000 . 000 
f .071 .000 .on .000 .091 .111 .156 .278 . 321 . 022 .OS6 . 028 . 000 .000 .000 F .250 . 000 . 000 . 000 .000 .000 . 000 
G .054 .000 .coo .000 .ooo . 000 .000 . 000 .036 .000 . 000 . 125 .000 .000 .ooo G .000 .000 . 000 . 000 .000 .000 1.000 
H . 000 .000 . 000 .000 .000 .000 .000 .000 .000 .000 .000 .OBJ .000 .000 .000 H . 750 1 .000 1.000 1 .000 .000 .000 . 000 
LOH• LOH* 
(N) JO 21 38 16 22 27 17 20 14 46 43 41 2 2 1 (N ) 6 5 3 12 6 ! 10 
. 000 . 000 .000 . 000 .000 .000 .000 .000 .000 .000 .000 .000 1 .000 1 .000 1.000 A .000 .000 .000 .000 . 000 .000 .200 
.000 .000 . 000 . 000 .coo . 000 . 000 .000 . 000 .000 .000 . 000 .000 .000 .coo 8 .000 .000 .000 .000 . 000 .000 .aoo 
.033 .095 . 000 .094 .068 . 000 .118 . 350 .250 .022 . 000 .012 . 000 . 000 .000 C .000 .000 .000 .000 . 000 .000 . 000 
.!33 .881 .868 .813 .932 .!33 .!24 . 5r.; . 500 .!26 .!49 .988 .000 . 000 .000 D 1.000 1.000 1.000 1.000 1.000 . 938 . 000 
.133 .024 .092 . 063 . 000 .019 .000 .000 . 143 .0!7 .061 . 000 .000 . 000 .000 E .000 .000 .000 .000 .000 .000 .000 
. 000 . 000 .039 . 000 . 000 .148 .059 . or.. . 107 . 054 .058 . 000 . 000 .000 .000 f .000 .000 .000 . 000 .000 .063 .000 
.000 .000 . 000 . 031 .000 . 000 . 000 . 000 .000 . 011 . 000 . 000 .000 .000 .000 G . 000 .000 .000 . 000 . 000 . 000 .000 
• 000 .000 . 000 .000 . 000 .000 .000 .000 .000 .000 .012 . 000 .000 .000 .000 • .000 . 000 . 000 . 000 .000 .000 .000 
········--·······-···· ·· · ············ · ··· ······· ····· ··· ----· · ··· · ··· -· · · ·-···· · ·· ····· · ·· · ···-········· ····· ·· · · ·· --
Mobi l it y of the a l leles in thtst loci does not follow tht 1l?"11bet 011c t ly . Actual mobility rank frar. fut to slow: 
PGH: ABCOEJftCH IL; HE: AHBCOIEJfK; PEP·1: ABCOElfCiM; HOH · 1: ABCOEFHI G; LOH: ABCOIEfGH. 
Population IICl"'onyTl\1 fo l low t1bl e 1· 3 . Host plant spec iu following •7 • 1rt U"ICtrt1ln In some Okl1hOl'M popu l at ion,. 
Table 1-7. X2 tests for differences in allele frequency among nymphal populations 
of bud and petiole gall makers, all loci. 
Contrast Contrast* 
number 
*. 
* * . 
* * *. 
NS: 
AO all four populations of bud galls 
(glabrous and hairy types) 
AOO all three populations of 
glabrous bud galls 
Al glabrous bud gall (MD) vs. 
glabrous bud gall (VA) 
A2 glabrous bud gall (MD) vs. 
glabrous bud gall (LA) 
A3 glabrous bud gall (MD) vs. 
hairy bud gall (OK) 
A4 glabrous bud gall (VA) vs. 
glabrous bud gall (LA) 
A5 glabrous bud gall (VA) vs. 
hairy bud gall (OK) 
A6 glabrous bud gall (LA) vs. 
hairy bud gall (OK) 
BO all 3 populations of petiole galls 
Bl petiole gall from MD(BB) vs. 
TX(BFL) 
B2 petiole gall from MD(BB) vs. 
AZ 
B3 petiole gall from TX(BFL) vs. 
AZ 
Population acronyms follow table 1-3. 
0.01 ~ p ~ 0.05. 
P < 0.005. 
non-s ignificant, p > 0.05. 
125 
Chi- degrees Signifi-
square of ca nee 
freedom level 
59.88 24 *** 
23.35 12 NS 
9.69 4 NS 
4.30 5 NS 
11.11 5 NS 
12.6 6 NS 
17.3 6 ** 
15.43 7 NS 
87.42 14 *** 
21.18 6 *** 
52.89 7 *** 
38.01 6 *** 
T able 1-8. Body size measurement of Pachypsylla gall makers from various gall types. Matrix at the 
bottom shows pairwise comparisons between means of leaf gall makers using Tukey-Kramer method 
(Sokal and R ohlf, 1981). The absolute pair average mean differences are given below the diagonal 
while minimum significant difference (MSD) value are given above the diagonal. Differences larger 
in absolute value than their MSD value are significant at the 0.05 level and are marked with an 
asterisk. 
Gall types Non-leaf galler Leaf galler 
Petiole Glabrous Disc Hairy Glabrous Star Blister 
gall bud gall nipple nipple gall gall 
gall gall gall 
Number of 14 15 42 59 54 99 98 
individuals 
Mean 6.02 3.25 3.83 3.34 3.27 2.98 2.65 
Body Minimum 5.50 2.90 3.10 2.70 2.70 2.4 1.9 
size Maximum 6.40 3.60 4.40 4.10 3.50 3.6 3.3 
Standard 0.07 0.06 0.05 0.05 0.03 0.03 0.04 
E rror 
Disc 
gall 
- 0.172 0.175 0.158 0.040 
Ranked Hairy 
gall nipple 0.489* - 0.161 0.140 0.140 
size gall 
Glabrous 
nipple 0.559* 0.070 - 0.144 0.143 
gall 
Star 
gall 0.85* 0.360* 0.291 * - 0.120 
Blistr 1.184* 0.695* 0.625* 0.344* -
gall 
126 
Table 1-9. X2 tests for allele frequency differences among populations of leaf gall makers, all loci. 
Contrast Contrast* 
number 
*· 
***· 
NS: 
co all 5 leaf gall types from GFV (adults) 
Cl blister vs. hairy nipple gall (GFV) 
C2 blister vs. glabrous nipple gall (GFV) 
C3 blister vs. star gall (GFV) 
C4 blister vs. disc gall (GFV) 
cs hairy vs. glabrous nipple gall (GFV) 
C6 hairy nipple vs. star gall (GFV) 
C7 hairy nipple vs. disc gall (GFV) 
C8 glabrous nipple vs. star gall (GFV) 
C9 glabrous nipple vs. disc gall (GFV) 
ClO star vs. disc gall (GFV) 
DO all 3 leaf gall types from GFV (nymphs) 
Dl blister vs. glabrous nipple gall (GFV) 
D2 blister vs. disc gall (GFV) 
D3 glabrous nipple vs. disc gall (GFV) 
El hairy nipple vs. blister gall (CMt) 
Fl glabrous nipple vs. star gall (MD) 
Gl blister galler (TX) on 
Ce/tis /aevigata vs. Ce/tis reticulata 
Population acronyms follow table 1-3. 
P < 0.005. 
non-significant, p > 0.01. 
127 
Chi- degrees Sign iii-
square of cance 
freedom level 
48.77 8 *** 
58.96 14 *** 
280.43 56 *** 
88.05 14 *** 
78.23 12 *** 
35.57 11 *** 
46.16 13 *** 
82.74 13 *** 
45.00 13 *** 
18.84 7 *** 
78.81 13 *** 
261.24 24 *** 
107.83 10 *** 
30.86 10 *** 
200.93 14 *** 
86.16 12 *** 
48.66 17 *** 
20.95 15 NS 
I 
Table 1-10 F . 
sampl ) f · r~quencies of major alleles in IDH-1 in adults (1991 
es rom blister galls versus hairy nipple galls. 
-
Great Falls Catoctin Mt. Wooster 
Allele Virginia Maryland Ohio 
Blister 
Hairy Hairy Hairy 
EB Nipple Blister Nipple Blister Nipple .86 .50 .73 .25 .75 .25 .05 .31 .27 .67 .13 .69 
128 
Table 1-11. Numbers and types of gall formed by progenies of individual females in rearing 
experiment. One large and one small female were caged on each seedling. Empty cells 
represent no galls of that type. 
PPOPU- HOST SEED- LARGE SIZE** SMALL SIZE** 
LATION (DATE OF LING FEMALE(> 3mm) FEMALE ( < 3mm) 
* RELEASE) # 
NIPPLE STAR NIPPLE STAR 
GALL GALL GALL GALL 
1 3 
Celtis 2 50 30 
tenuifolia 
(5/13) 3 8 14 
4 15 29 
NAL 5 14 
31 8 
Cell is 32 9 
tenuifolia 
(5/24) 
33 8 
34 
11 
Celt is 12 
tenuifolia 13 15 15 
(5/17) 
14 6 
BB 15 13 8 
21 10 10 
Ce I tis 22 
occidental is 23 
(5/23) 24 
* NAL: National Agricultural Library, Beltsville, Maryland; 
BB: Branchville Rd., Berwyn, ~a~yland: . . 
2 B d · I d trong assoc1at1on with type of gall made (2X2 contmgency test, X o y size revea e s ** 
== 5.5, p < 0.01) . 
129 
Table 1-12. Head/shaft ratio of the terminal segment of male aedeagus in blister gall makers of 
Pachyp!iylla on different hosts. 
* 
Head/shaft ratio 
Host Number of 
individuals Mean Minimum Maximum Standard 
Error 
Celtis laevigata 
(TX)* 9 a.so 0.35 0.65 0.038 
Ceilis reticulata 
(TX) * 12 0.33 0.24 0.44 0.029 
Ccltis rcti cu lata 
(AZ) 4 0.33 0.32 0.37 O.D28 
Celt is 
occidentalis 10 0.33 0.31 0.37 0.007 
(MD) 
The ratios between blister gallers from C. laevigata and C. reticulata from Texas differed from 
each other significant ly (t-test, t = 3.59, df = 19, P < 0.01). 
130 
Table 2-1 
• •. . •• .. . n 1992 "" 1993 " ny,,,h• fr~ d, ffecen< cell ''""""' ,m ;n sadM 
. . . . . . . . . ..... : '°""" fo' " ll <ype• and "''" l a< ; '"' f o ll"' 1abl , 1-3 go ll ""' collected .i Allele frequencies of .
i:::, 1i· ,)::, ... :i: .... -::: ..... :i:· .... ;······ ',""''"; ·······;·······;······;;-: ..... 1,::.. · ;; 
Cell po, • mo= (CM<) (CM<) (CM<) ( L"9 LNO L"9 LS L"9 L'9 l'9 lS center side NAL) (NAU (NAU (NAL) (BB) (BB) (BB) (BB) 
- ---- ------- mono mono center side mono mono center side mono 
------------------------------------------------------
-. -.. ---- . ---
PGH 
(N) 
A 
B 
C 
D 
E 
F 
G 
H 
I 
PGI 
(N) 
A 
B 
C 
D 
E 
F 
G 
GPDH 
( N} 
A 
B 
C 
D 
HE 
(N) 
A 
B 
C 
D 
E 
F 
FUM 
(N) 
A 
B 
C 
D 
E 
F 
PEP -1 
(N) 
A 
B 
C 
D 
E 
F 
FDp 
(N) 
A 
B 
lPI 
(N) 
A 
B 
C 
D 
E 
16 
.000 
.ooo 
.031 
.344 
.031 
.563 
.ooo 
.031 
.ooo 
16 
.ooo 
.ooo 
.938 
.ooo 
.063 
.ooo 
.ooo 
12 
.ooo 
.ooo 
1.000 
.ooo 
16 
.ooo 
.ooo 
.ooo 
1 .ooo 
.ooo 
.ooo 
16 
.ooo 
.ooo 
.ooo 
.969 
.ooo 
.031 
7 
.ooo 
.429 
.571 
.ooo 
.ooo 
.ooo 
12 
.ooo 
1 .ooo 
12 
.ooo 
1.000 
.ooo 
.ooo 
.ooo 
14 
.000 
.000 
.000 
• 179 
.000 
.607 
.000 
.214 
.ODO 
14 
.000 
.000 
.964 
.000 
.036 
.000 
.000 
11 
.000 
.000 
1.000 
.000 
14 
.000 
.000 
. 000 
.964 
.000 
.036 
14 
.000 
.000 
. 000 
1.000 
.000 
.000 
5 
. 100 
.200 
.600 
. 000 
.100 
.000 
11 
.000 
1.000 
11 
.000 
1.000 
21 
.000 
.ODO 
.000 
.024 
.000 
.786 
.000 
.119 
.071 
21 
.000 
.000 
1 .000 
.000 
.000 
.000 
.000 
18 
. 000 
.000 
1.000 
.000 
21 
.000 
.000 
.000 
.286 
.119 
.595 
21 
.000 
.000 
.000 
.905 
.ODO 
.095 
12 
.000 
.250 
.625 
.042 
.083 
.000 
18 
.ODO 
1.000 
18 
.000 
1.000 
18 
.ooo 
.000 
.ODO 
.222 
.028 
.639 
.028 
.083 
.000 
18 
.000 
.028 
.972 
.000 
.000 
.000 
.000 
17 
.000 
.029 
.971 
.000 
18 
.056 
.ODO 
.ODO 
.917 
.ODO 
.028 
17 
.000 
.000 
.ooo 
.912 
.000 
.088 
14 
.000 
.000 
1.000 
.000 
.ooo 
.000 
8 
.000 
1.000 
12 
.ooo 
1.000 
17 10 13 
.ODO .ooo .ooo 
.000 .ooo .000 
.ODO .ODO .ODO 
.176 .zoo .ooo 
.088 .000 .000 
.647 .700 .923 
.029 .ooo .ooo 
.059 .100 .077 
.ooo .ooo .000 
17 10 13 
.ooo .ooo .ooo 
.ODO .ooo .ooo 
.824 .900 1.000 
.088 .ooo .ODO 
.088 .100 .ooo 
.000 .ODO .000 
.000 .ooo .ooo 
17 10 13 
.ooo .ooo .ooo 
. 088 .250 .ooo 
.912 . 750 1.000 
.000 .ooo .ooo 
20 13 16 
.075 .038 .ooo 
.ooo .ooo .ooo 
.ooo .ooo .ooo 
.900 .885 .531 
.ooo .ooo 
.031 
.025 .077 .438 
17 10 
13 
.ooo .ooo 
.ODO 
.ooo .050 
.ooo 
.000 .ooo 
.ooo 
.971 .950 
1.000 
.ooo .ooo 
.ooo 
.029 .ODO .ooo 
9 4 
7 
.ooo .ooo 
.ooo 
.111 .125 
, 143 
.778 .875 
.857 
.ooo .ooo 
.ooo 
.111 ,000 
.ooo 
.ooo .ooo 
.ooo 
14 7 
10 
.ODO .ooo 
.ooo 
1,000 1.000 
1.000 
14 7 
10 
.ooo .ooo 
.ooo 
1.000 1.000 
1.000 
.ooo .ooo 
11 
.ooo 
.ooo 
.ODO 
.045 
.ooo 
.955 
.ooo 
.ODO 
.ooo 
13 
.000 
.ooo 
.846 
.ooo 
.077 
.ooo 
.OT! 
13 
.077 
.077 
.846 
.ooo 
16 
.ODO 
.ooo 
.ooo 
.844 
.031 
.125 
13 
.ooo 
.ooo 
.ooo 
.962 
.ooo 
.038 
11 
.ooo 
, 182 
.818 
.ooo 
,000 
.ooo 
11 
.ooo 
1.000 
8 
.ooo 
1.000 
.ooo 
.ooo 
3 
.ooo 
.ODO 
.000 
.ODO 
.167 
.667 
.000 
.167 
.000 
3 
.ooo 
.ooo 
.833 
.ODO 
.167 
.ODO 
.ooo 
3 
.ooo 
.ooo 
1.000 
.ooo 
3 
.ooo 
.ooo 
.ooo 
1.000 
.ooo 
.ooo 
3 
.ooo 
.ooo 
.ooo 
1.000 
.ooo 
.ooo 
3 
.ooo 
.ooo 
,667 
.ooo 
, 167 
.167 
3 
.ooo 
1.000 
3 
.ooo 
1.000 
.ooo 
.ooo 
3 
.ODO 
.ODO 
.000 
.ooo 
.000 
1.000 
.000 
.ooo 
.ooo 
3 
.ooo 
.ooo 
1.000 
.ooo 
.ooo 
.ooo 
.ooo 
3 
.ooo 
.ooo 
1 .ooo 
.ooo 
3 
.ooo 
.ooo 
.ooo 
1.000 
.ooo 
.ooo 
3 
.ooo 
.ooo 
.ooo 
1.000 
.ooo 
.ooo 
2 
.ooo 
.ooo 
1.000 
.ooo 
.ooo 
,000 
3 
.000 
1.000 
3 
.ooo 
1.000 
.ooo 
.ooo 
6 
.ODO 
.000 
.ODO 
.000 
.ODO 
.833 
.000 
.167 
.000 
6 
.000 
.ooo 
1.000 
.000 
.ooo 
.ODO 
.000 
6 
.ODO 
.ooo 
.917 
.083 
6 
.000 
.ODO 
.000 
.667 
.000 
.333 
6 
.ooo 
.000 
.ooo 
1.000 
.ooo 
.000 
5 
,000 
.ooo 
1.000 
.ooo 
.ooo 
.ooo 
6 
,000 
1.000 
6 
.ODO 
.917 
.ooo 
.000 
.083 
11 
.000 
.000 
.000 
.136 
.045 
.773 
.045 
.000 
.000 
11 
.ooo 
.000 
.955 
.ODO 
.000 
.045 
.000 
11 
.091 
.000 
.909 
.ooo 
11 
.000 
.ooo 
.000 
.955 
.045 
.ooo 
11 
.000 
.000 
.ooo 
.909 
.ooo 
.091 
11 
,000 
.045 
.955 
.ooo 
.ooo 
.ooo 
9 
,000 
1 .ooo 
8 
.063 
.938 
.ooo 
.ooo 
, 000 
.000 .000 .ODO .ooo 
. 000 . 000 .ooo .ooo 
.ooo .ooo 
.000 .000 .000 .ooo 
.ooo .ooo 
.ooo ,000 
.ooo 
-- ---- ------ ----------- ---------------------- ---- ------- ---- -- -------- -------- -- --- -
131 
Si 
Table 2-1. (cont.) 
-------------------------------------------------------------------------------------------------------
Population 
----------------------------------------------------------------------------------------------
1 2 3 4 5 6 7 8 9 
10 11 12 
Gall type LNh LNh LNh LB LNg LNg LNg LS LNg 
LNg LNg LS 
Locality (CMt) (CMt) (CMt) (CMt) (NAL) (NAL) (NAL) (NAL) (BB) (BB) (BB) (BB) 
Cell pos." momo center side mono mono center side mono mono 
center side mono 
-------- -- ---------------------------------------------------------------------------------------------
G3PDH 
(N) 12 11 18 10 14 7 10 11 3 3 6 11 
A 
.000 .000 .000 .000 .ooo .000 .000 .000 .000 
.000 .000 .000 
B 1.000 1.000 1.000 1.000 1.000 1 .000 1.000 1.000 1.000 1.000 
1.000 1.000 
IDH-1 
(N) 16 14 18 18 20 13 16 16 3 3 6 11 A 
.ooo 
.000 .000 .000 .000 .000 .000 .000 .000 .000 
.000 .000 
B 
.ooo 
.000 .000 .000 .000 .000 .000 .000 .000 .000 
.000 .000 
C .ooo 
.000 .000 .000 .000 .000 .000 . 188 .000 .000 
.000 .227 
D 
.250 
.214 .389 .972 . 125 .077 .031 .219 .000 .167 
.000 .182 
E 
.750 
.786 .611 .028 .850 .923 .938 .500 .833 .833 
.917 .500 
F 
.ooo 
.000 .000 .000 .025 .000 .000 .094 .167 .000 .000 
.091 
G 
.ooo 
.000 .000 .000 .000 .000 .031 .000 .000 .000 
.083 .000 
IDH-2 
(N) 16 13 21 15 20 13 16 16 3 3 6 
11 
A 
.ooo 
.000 .000 .000 .000 .000 .000 .000 .000 .000 
.000 .000 
B 
.ooo 
.000 .000 .000 .000 .000 .000 .000 .000 .000 .000 
.000 
C 
.ooo 
.000 .000 .000 .000 .077 .000 .000 .000 .000 .000 
.000 
D 1.000 1.000 1 .000 1.000 .950 .923 1 .000 1 .000 1 .000 1.000 1 .000 
1.000 
E 
.ooo 
.000 .000 .000 .050 .000 .000 .000 .000 .000 .000 .000 MDH· 1 
(N) 15 13 21 13 17 10 13 8 3 3 6 8 A 
.ooo 
.000 .000 .059 . 100 .000 .000 .000 .000 .000 . 125 .000 B 
.067 
.077 .346 .206 .050 . 115 .250 . 167 . 167 .250 . 188 C .024 
.067 
.000 .048 .000 .ooo .000 .000 .000 .000 .000 .000 .000 D 
.867 
.885 .654 .735 . 750 .654 . 750 .833 .667 .583 .688 E .405 
.ooo 
.000 . 119 .000 .000 .000 . 115 .000 .000 .000 .000 .000 F 
G 
.ooo 
.038 .381 .000 .ooo .050 .077 .000 .000 .167 .083 .000 
MDH-2 
.ooo 
.000 .024 .000 .000 .050 .038 .000 .000 .000 .083 .000 
(N) 12 14 14 20 13 16 16 3 3 6 11 18 A 
.ooo 
.000 .ooo .000 .000 .000 .000 B .000 .000 
.000 .000 .000 
C 
.ooo 
.000 .000 .000 .000 .000 .000 .000 .000 .000 .000 .000 
6PGDH 1 .ooo 1.000 1 .000 1.000 
1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 
00 15 13 16 20 13 16 16 3 3 6 11 A 21 
B 
.ooo 
.000 .024 .000 .ooo .000 .000 .000 .000 .000 .000 .000 
C .033 .000 .024 .000 
.ooo .000 .000 .000 .000 .000 .000 .000 
D 
.033 
.000 .000 .000 .ooo .000 .000 .000 .000 .000 .000 .000 
E 
.867 
.808 .952 .813 .975 1 .000 .781 .875 1.000 1.000 1.000 .864 
F 
.ooo 
.000 .000 .031 .ooo .000 .094 .031 .000 .000 .000 .000 
G .067 .077 .000 . 156 
.025 .000 .125 .094 .000 .000 .000 . 136 
LOH .ooo . 115 .000 .000 
.ooo .000 .000 .000 .000 .000 .000 .000 
(N} 16 19 13 16 16 3 3 6 
A 14 21 17 
11 
B 
.ooo 
.000 .000 .000 .ooo .000 .000 .000 .000 .000 .000 .000 
C .ooo .000 .000 .000 
.ooo .000 .000 .000 .000 .000 .000 .000 
D .031 .036 .095 . 118 .ooo 
.000 .063 .000 .000 .000 .083 .000 
E 
.875 
.786 .881 .824 .763 .962 .938 .813 1 .000 1.000 .917 .864 
F .094 .179 .024 .000 .158 
.038 .000 .000 .000 .000 .000 .045 
.ooo 
.000 .000 .059 .079 .000 .000 .188 .000 .000 .000 .091 
---------------------------------
-----------------
--------- --------- -- -- -- - --
" Cell Pos ition within an individual gall. 
132 
Table 2-2. x2 . >nak ,s in th tests fo, allele frequency diffe,ences between nymphal populatwns of leaf gall 
e same cell positions all loci 
' 
=--
Contrast Contrast* 
Chi- degrees 
Signifi-
number 
square of 
cance 
freedom level 
6.69 10 
NS 
Al Hairy nipple gall: 
r--- mono vs. center cell (CMt) 13 NS 14.90 
A2 Glabrous nipple gall: 
t---~mono vs. center cell (NAL) 5.60 5 NS 
A3 Glabrous nipple gall: 
t--- mono vs. center cell (BB) 
Bl Glabrous nipple gall (mono + center cells): 
35.69 33 
NS 
---
_l"IAL vs. BB 
B2 Glabrous nipple gall (side cell): 
10.28 9 
NS 
---· 
NAL vs. BB 10.67 11 
NS 
B3 Star gall (mono cell): 
~ NAL vs. BB 
*· N Populat" S: n n . 1?n acronyms follow table 1-3. 
-sign di cant, p > 0.01. 
133 
Tab~e 2-3. A . . positions _lle_le fr~uenc1es (after pooling) of n~s from different cell 
allele tr' ~ithin :,,ar1ous gal l types collected in 1992 and 1993. Major 
Under linedeq ency differences between side cell and center cell nymphs are 
• Acronyms for gall types and populations follow Table 1-3. 
-- -- -- ---- ----------------------- ------- ------- -----
Population 
----·--
---- ------------------------------------ ---------- -5 6 
Gall t ype 
Locality 
Cel l pos .* 
---
PGM 
(N) 
C 
D 
E 
F 
G 
H 
I 
PG! 
(N) 
C 
D 
E 
F 
G 
GPDH 
(N) 
A 
B 
C 
D 
ME 
CN) 
A 
D 
E 
F 
FUM 
(N) 
B 
D 
F 
PEP - 1 
(N) 
A 
B 
C 
D 
E 
F 
FDp 
(N) 
A 
B 
TPI 
CN) 
A 
B 
E 
LNh 
(CMt) 
m+-c 
30 
.017 
.267 
.017 
.583 
.000 
.117 
.000 
30 
.950 
.000 
. 050 
. 000 
.000 
23 
.000 
.000 
1.000 
.000 
30 
. 000 
. 983 
. 000 
.017 
30 
.000 
.983 
.017 
12 
.042 
.333 
.583 
. 000 
. 042 
.000 
23 
.000 
1.000 
23 
. 000 
1.000 
.000 
2 
LNh 
(CMt) 
side 
21 
.000 
.024 
.000 
.786 
.000 
. 119 
.071 
21 
1. 000 
, 000 
.000 
.000 
.000 
18 
.000 
.000 
1.000 
.000 
21 
.000 
.:1§2 
• 119 
.595 
21 
.000 
.905 
.095 
12 
.000 
.250 
.625 
.042 
. 083 
. 000 
18 
.000 
1. 000 
18 
.000 
1.000 
.000 
3 4 LB LNg LNg LS 
(CMt) (NAL+BB) (NAL+BB) (NAL+BB) 
mono m+-c side mono 
18 
.ooo 
.222 
,028 
.639 
.028 
. 083 
.000 
18 
.972 
.ooo 
.000 
.ooo 
,000 
17 
.ooo 
.029 
.971 
.000 
18 
. 056 
.917 
.ooo 
_,.91§ 
17 
.ooo 
.912 
.088 
14 
.ooo 
.000 
1.000 
.ooo 
.ooo 
.ooo 
8 
, 000 
.000 
12 
.ooo 
1.000 
.ooo 
33 
.ooo 
.152 
.061 
.697 
.015 
.076 
.ooo 
33 
.864 
.045 
.091 
.ooo 
.ooo 
33 
,000 
• 121 
.879 
.ooo 
39 
.051 
.,_21.Q 
.ooo 
~ 
33 
.015 
.970 
.015 
18 
.ooo 
.083 
.806 
.ooo 
.083 
, 028 
27 
.ooo 
.ooo 
27 
.ooo 
1.000 
.ooo 
134 
19 
.ooo 
,000 
.ooo 
.895 
.ooo 
. 105 
.ooo 
19 
1.000 
.ooo 
.ooo 
,000 
.ooo 
19 
.ooo 
.ooo 
.974 
.026 
22 
.ooo 
.:2.@ 
.023 
~ 
19 
.ooo 
1.000 
.ooo 
12 
.ooo 
,083 
.917 
.ooo 
.ooo 
.ooo 
16 
.ooo 
.ooo 
16 
.ooo 
,969 
.031 
22 
.ooo 
.091 
.023 
.864 
.023 
.ooo 
.ooo 
24 
.896 
.ODO 
. 042 
.021 
.042 
24 
,083 
.042 
.875 
,ODO 
27 
.ooo 
;.§§2 
,037 
.,_W. 
24 
.ooo 
.938 
.063 
22 
.ooo 
. 114 
,886 
.ooo 
.ooo 
.ooo 
20 
.ooo 
.ooo 
16 
.031 
.969 
.ooo 
Table 2-3. (cont.) 
----- --- ------------- --- --------- -- ---- --------- -------------------
Gall type 
Locality 
Ce ll pos.* 
G3PDH 
(N) 
A 
B 
IDH-1 
(N) 
C 
D 
E 
F 
G 
IDH-2 
(N) 
C 
D 
E 
MDH-1 
(N) 
A 
B 
C 
D 
E 
F 
G 
MDH - 2 
(N) 
A 
B 
C 
6PGDH 
(N) 
A 
B 
C 
D 
E 
F 
G 
LDH 
(N) 
C 
D 
E 
1 
LNh 
(CMt) 
m+c 
23 
.000 
1.000 
30 
.000 
.233 
. 767 
.000 
.000 
29 
.000 
1.000 
.000 
28 
.000 
.071 
.036 
.875 
.000 
.018 
.000 
26 
.000 
.000 
1.000 
28 
.000 
.018 
.018 
.839 
.000 
. 071 
.054 
30 
.033 
. 833 
. 133 
.000 
Population 
2 3 
LNh LB 
(CMt) (CMt) 
side mono 
18 
.000 
1.000 
18 
.000 
.389 
.611 
.000 
.000 
21 
.000 
1.000 
.000 
21 
.000 
.024 
.048 
. 405 
. 119 
.381 
.024 
18 
.000 
.000 
1.000 
21 
.024 
.024 
.000 
.952 
.000 
.000 
.000 
21 
.095 
.881 
.024 
.000 
10 
.000 
1.000 
18 
.000 
.972 
.028 
.000 
.000 
15 
.000 
1 .000 
.000 
13 
.000 
.346 
.000 
.654 
.000 
.000 
.000 
14 
.000 
.000 
.000 
16 
.000 
.000 
.000 
.813 
.031 
. 156 
.000 
17 
.118 
.824 
.000 
.059 
4 5 6 
LNg LNg LS 
(NAL+BB) (NAL+BB) (NAL+BB) 
m+c side mono 
27 
.000 
1.000 
39 
.000 
• 103 
.872 
.026 
.000 
39 
.026 
.949 
.026 
33 
.061 
• 152 
.000 
.742 
.000 
.030 
.015 
39 
.000 
.000 
.000 
39 
.000 
.000 
.000 
.987 
.000 
.013 
.000 
38 
.000 
.868 
.092 
.039 
16 
.000 
1.000 
22 
.000 
.023 
.932 
.000 
.045 
22 
.000 
1.000 
.000 
19 
.000 
.158 
.000 
.632 
.079 
.079 
.053 
22 
.000 
.000 
.000 
22 
.000 
.000 
.000 
.841 
.068 
.091 
.000 
22 
.068 
.932 
.000 
.000 
16 
22 
.000 
1.000 
27 
.204 
.204 
.500 
.093 
.000 
27 
.000 
1.000 
. 000 
.063 
.219 
.000 
.719 
.000 
.000 
.000 
27 
.000 
.000 
.000 
27 
.000 
. 000 
.000 
.870 
.019 
.111 
.000 
27 
.000 
.833 
.019 
. 148 
-------------------------------------------------------------------
* Ce ll pos ition within an individual gall: m or mono= mono-cell; c= center 
cell; s ide= s ide cell. 
135 
r 
Table 2-4. X2 tests for allele frequency differences between nymphs in different 
cell positions within the same gall type, side cell nymphs and other co-occurring 
mono cell gall types, and side cell nymphs from different gall types, all loci. 
Contrast Contrast* 
number 
Cl Hairy nipple gall (CMt): 
(mono + center) vs. side cell 
C2 Glabrous nipple gall (NAL+ BB): 
( mono + center) vs. side cell 
Dl CMt: Hairy nipple gall (side cell) 
vs. Blister gall (mono cell) 
D2 NAL+ BB: Glabrous nipple (side 
cell) vs. Star gall (mono cell) 
E l Side cell: Hairy nipple gall vs. 
Glabrous nipple gall 
* 
*** 
Population acronyms follow table 1-3. 
p < 0.005. 
136 
Chi- degrees Signifi-
square of cance 
freedom level 
87.95 13 ** * 
63.87 15 *** 
106.87 16 *** 
58.36 14 *** 
53.78 13 *** 
Table 2-5. Number and kinds of galls formed by progenies of single females in 
rearing experiments testing for inquilinism of side cell individuals in glabrous nipple 
gall and star galls from two Maryland populations. 
* 
** 
POP UL- HOST** SEED- BROWN GREEN 
A TION* (DATE OF LING ABDOMEN~ ABDOMEN~ 
RELEASE) # (CENTER CELL) (SIDE CELL) 
SIN- MULTI- SIN- MULTI 
GLE PLE GLE PLE 
CELL CELL CELL CELL 
GALL GALL GALL GALL 
1 3 
Ce/tis 2 54 
tenwfolia 3 22 (5/13) 
4 18 
NA L 5 14 
31 8 
Ce/tis 32 6 
tenuifolia 33 7 (5/24) 
34 
11 
Ce/tis 12 
te11111f olia 13 6 
(5/17) 
14 6 
BI3 15 8 
21 5 
Ce/tis 22 
occidentalis 23 
(5/23) 
24 
NA L: Natio nal Agricultural Library, Beltsville, Maryland; 
13B: Bra nchville Rd., Berwin, Maryland. 
C.t.: Ce/tis te1111ifolia; C.o.: C. occidentalis. 
*** Green abdomen female died early in experiment. 
137 
GREEN+ BROWN 
ABDOMEN~~ 
(CENTER+SIDE) 
SINGLE MULTI-
CELL PLE 
GALL CELL 
GALL 
10 6 
1 14 
19*** 
8 5 
7 5 
20 18 
12** 
9 6 
~ 
\..,.) 
00 
Table 3-1. Character matrix for all evidences: morphological characters (1-33); allozyme data (34-48); 
karyotype (49-50); and life history (51 -52) coded from Table 1-6. Each locus is treated as single 
character. 
Pachy4.ALL (alle0.2) 1 1 2 3 4 5 6 7 8 9 1 0 11 12 1 3 
1 Hairy Nipple Gal l (center cel l) 3 1 2 1 0 1 0 0 1 1 1 1 1 
2 Hairy Nipple Gall (side cell) 3 1 2 1 0 1 0 0 1 1 1 1 1 
3 Glabrous Nipple Gall (center ce 3 1 2 1 0 1 0 0 1 1 1 1 1 
4 Glabrous Nipple Gall (side cell 3 1 2 1 0 1 0 0 1 1 1 1 1 
5 Star Gall (center cell) 3 1 2 1 0 1 0 0 1 1 1 1 1 
6 Blister Gall (center cell) MO 3 1 2 1 0 1 0 0 1 1 1 1 1 
7 Blister Gall (center cell) TX 3 1 2 1 0 1 0 0 1 1 1 1 1 
8 1 Oise Gall (center cell) 3 1 2 1 0 1 0 0 1 1 1 1 1 
9 I Petiole Gall 3 1 2 0 0 1 0 1 1 0 2 1 1 
10 Glabrous Bud Gall 2 1 \ 1 0 0 1 0 1 1 1 2 1 1 
11 \ Hairy Bud Gall 2 I 1 1 0 0 1 0 1 1 ? 2 1 1 
12 \ Twig Gall 2 1 1 0 0 1 0 1 1 ? 2 1 1 
13 \ Tetragonocephala 1 I 0 I 0 0 1 0 1 1 0 1 0 0 0 
1 4 \ Celtisaspis 
' 
0 I 0 I 0 0 I 2 I 0 1 I 1 0 I 1 0 I 0 1 
\1 
:i 
.\ I 
\ 
-\, 
\ 
>--' 
w 
\0 
Table 3-1 (cont.). 
Pachy4.ALL (alle0.2) 2 
1 Hairy Nipple Gall (center cell) 
2 Hairy Nipple Gall (side cell) 
3 Glabrous Nipple Gall (center ce 
4 Glabrous Nipple Gall (side cell 
5 Star Gall (center cell) 
6 Blister Gall (center cell) MD 
7 Blister Gall (center cell) TX 
8 Disc Gall (center cell) 
9 Petiole Gall 
10 G\abrous Bud Gall 
11 Hairy Bud Gall 
12 Twig Gall 
13 Tetragonocephala 
14 Celtisaspis 
Pachy4.ALL (alle0.2) 3 
1 Hairy Nipple Gall (center cell) 
2 Hairy Nipple Gall (side cell) 
3 Glabrous Nipple Gall (center ce 
4 Glabrous Nipple Gall (side cell 
5 Star Gall (center cell) 
6 Blister Gall (center cell) MD 
7 Blister Gall (center cell) TX 
8 Disc Gall (center cell) 
9 Petiole Gall 
1 0 Glabrous Bud Gall 
11 Hairy Bud Gall 
12 Twig Gall 
13 Tetragonocephala 
14 Celtisaspis 
14 15 
1 0 
1 0 
1 0 
1 0 
1 0 
0 0 
0 0 
1 0 
1 1 
1 0 
1 0 
1 0 
0 0 
0 0 
27 28 
2 0 
2 0 
2 0 
2 0 
2 0 
0 0 
0 0 
2 0 
0 0 
0 1 
0 1 
1 1 
0 0 
0 0 
' ' 16 17 18 19 
0 1 1 1 
1 1 1 1 
0 1 1 1 
1 1 1 1 
0 0 0 1 
0 1 0 1 
0 1 0 1 
0 1 1 1 
0 0 1 1 
0 0 0 1 
0 0 0 1 
0 0 0 1 
0 0 0 0 
0 0 0 0 
29 30 31 32 
1 1 0 0 
1 1 0 0 
1 1 0 0 
1 1 0 0 
1 1 0 0 
1 1 0 0 
1 1 0 0 
1 1 0 0 
1 1 1 1 
1 1 1 0 
1 1 1 0 
1 1 1 1 
0 0 - -
I 0 0 - I 
20 21 22 23 24 25 26 
0 0 0 0 0 1 1 
0 0 0 0 0 1 1 
0 0 0 0 0 1 1 
0 0 0 0 0 1 1 
0 0 0 0 0 1 1 
0 0 0 0 0 1 1 
0 0 0 0 0 1 1 
0 0 0 0 0 1 1 
0 0 0 0 1 1 2 
0 0 1 0 1 1 2 
0 0 1 0 1 1 2 
0 0 1 0 1 1 2 
1 1 0 1 0 0 0 
1 1 0 1 2 0 0 
I 33 34 I 35 I 36 37 I 38 I 39 
PGM PGI GPCH /vE FUM PE Pl 
0 D/ F C C D D 8 /C 
0 F C C D/ F D 8 / C 
0 F C C D D C 
0 F C C 0 / F D C 
0 F C C D D C 
0 0/F C C D D C 
0 D/F C C D D C 
0 D/F C C D D 8 / C 
0 J K F A E/G 8 /C 
1 J/K J G J/K/Q C E 
1 J/K J G Q C E 
0 K H D H C F 
0 B J / Q C B F Q 
0 I L I F C I C I F G 
I--'-
~ 
0 
Table 3-1 (cont.). 
Pachy4.ALL (alle0.2) 4 40 41 
RP TPI 
1 Hairy Nipple Gall (center cell) B B 
2 Hairy Nipple Gall (side cell) B B 
3 Glabrous Nipple Gall (center ce B B 
4 Glabrous Nipp.le Gall (side cell B B 
5 Star Gall (center cell) B B 
6 Blister Gall (center cell) MD B B 
7 Blister Gall (center cell) TX B B 
8 Disc Gall (center cell) B B 
9 Petiole Gall B D 
1 0 Glabrous Bud Gall B B/E 
1 1 Hairy Bud Gall B E 
12 Twig Gall B C 
13 T etragonocephala B B 
14 1 Celtisaspis \ A I B 
Pachy4.ALL (al\e0.2) 5 1 53 I 
ovioosi 
1 Hairy Nipple Gall (center cell) 1 
2 Hairy Nipple Gall (side cell) 1 
3 Glabrous Nipple Gall (center ce 1 
4 Glabrous Nipple Gall (side eel\ 1 
5 Star Gall (center eel\) 1 
6 Blister Gall (center cell) MD 1 
7 Blister Gal\ (center eel\) TX 1 
8 Disc Gall (center cell) 1 
9 Petiole Gall 0 
10 Glabrous Bud Gall 2 
1 1 Hairy Bud Gall 2 
12 Twig Gall 2 
13 Tetragonocephala ? 
14 1 Celtisaspis I ? 
I 42 I 43 
G3PDH IDH1 
B D/ E 
B D/ E 
B E 
B E 
B CID/E 
B D 
B D 
B E 
B D 
B E 
B E 
B C/F 
B F 
I A I A/B 
I 
44 45 46 47 48 49 50 51 52 
IDH2 MDH1 MDH2 6PGD l.DH M-KAR F-KAm voltine overwi 
D D C D D 5 6 1 2 
D D/ F C D D 5 6 1 2 
D D C D D 5 6 1 2 
D D C D D ? ? 1 2 
D BID C D D ? ? 1 2 
D BID C D D 5 6 1 2 
D D C D D ? ? 1 2 
D D C D D ? ? 1 2 
G D/Q C B A 3 4 1 1 
G D C H D 2 2 1 1 
G D C H D ? ? 1 1 
G H C C D 5 6 1 1 
B G A A/B D 5 6 2 ? 
A G I A I G A/B ? I ? 1 / 2 0 
...... 
.i::,. 
...... 
Table 3-2. Character matrix for all evidences: morphological characters (1-33); allozyme data (34-130); 
karyotype (131-132); and life history (133-135) coded from Table 1-6. Each allele is treated as single 
character. 
Pachy4.ALL (s.allele) 1 I I 1 2 3 4 5 6 7 8 9 10 11 12 13 
1 Hairy Nipple Gall (center cell) 3 1 2 1 0 1 0 0 1 1 1 1 1 
2 Hairy Nipple Gall (side cell) 3 1 2 1 0 1 0 0 1 1 1 1 1 
3 Glabrous Nipple Gall (center ce 3 1 2 1 0 1 0 0 1 1 1 1 1 
4 Glabrous Nipple Gall (side cell 3 1 2 1 0 1 0 0 1 1 1 1 1 
5 Star Gall (center cell) 3 1 2 1 0 1 0 0 1 1 1 1 1 
6 Blister Gall (center cell) MD 3 1 2 1 0 1 0 0 1 1 1 1 1 
7 Blister Gall (center cell) TX 3 1 2 1 0 1 0 0 1 1 1 1 1 
8 Disc Gall (center cell) 3 1 2 1 0 1 0 0 1 1 1 1 1 
9 Petiole Gall 3 1 2 0 0 1 0 1 1 0 2 1 1 
10 Glabrous Bud Gall 2 1 1 0 0 1 0 1 1 1 2 1 1 
11 Hairy Bud Gall 2 1 1 0 0 1 0 1 1 ? 2 1 1 
12 Twig Gall 2 1 1 0 0 1 0 1 1 ? 2 1 1 
13 Tetrogonocephala 1 0 0 0 1 0 1 1 0 0 0 0 0 
1 4 \ Cel\isaspis \ 0 I 0 I 0 0 2 0 I 1 I 1 \ 0 0 I 0 I 0 1 
p achy4.ALL (s.allele) 2 1 15 16 I 17 18 ' 19 ' 20 21 22 23 24 25 26 27 
1 \ Hairy Nipple Gall (center cell) 0 I 0 1 1 1 0 0 0 0 0 1 1 2 
2 \ Hairy Nipple Gall (side cell) 0 I 1 I 1 \ 1 1 0 0 0 0 0 1 1 2 
3 \ G\abrous Nipple Gall (center ce 0 I 0 I 1 I 1 \ 1 0 0 0 0 0 1 1 2 
4 \ Glabrous Nipple Gall (side cell 0 I 1 1 \ 1 l 1 0 0 0 0 0 1 1 2 
5 \ Star Gall (center cell) 0 0 0 I 0 1 0 0 0 0 0 1 1 2 
6 Blister Gall (center cell) MD 0 0 1 0 1 0 0 0 0 0 1 1 0 
7 Blister Gall (center cell) TX 0 0 1 0 1 0 0 0 0 0 1 1 0 
8 Disc Gall (center cell) 0 0 1 1 1 0 0 0 0 0 1 1 2 
9 Petiole Gall 1 0 0 1 1 0 0 0 0 1 1 
2 0 
10 G\abrous Bud Gall 0 0 0 0 1 0 0 1 0 1 1 2 
0 
11 Hairy Bud Gall 0 0 0 0 1 0 0 1 0 1 1 2 0 
12 Twig Gall 0 0 0 0 1 0 0 1 0 1 1 2 1 
13 Tetrogonocephala 0 0 0 0 0 1 1 0 1 0 0 0 
0 
14 \ Celtisaspis I 0 \ 0 \ 0 0 I 0 1 1 0 1 2 0 0 0 
14 
1 
1 
1 
1 
1 
0 
0 
1 
1 
1 
1 
1 
0 
0 
28 
0 
0 
0 
0 
0 
0 
0 
0 
0 
1 
1 
1 
0 
0 
........ 
-i:,.. 
N 
Table 3-2 (cont.) 
Pachy4.ALL (s.allele) 3 
1 Hairy Nipple Gall (center cell) 
2 Hairy Nipple Gall (side cell) 
3 Glabrous Nipple Gall (center ce 
4 Glabrous Nipple Gall (side cell 
5 Star Gall (center cell) 
6 Blister Gall (center cell) MD 
7 Blister Gall (center cell) TX 
8 Disc Gall (center cell) 
9 Petiole Gall 
10 Glabrous Bud Gall 
11 Hairy Bud Gall 
12 Twig Gall 
13 Tetrogonocephala 
14 Celtisaspis 
Pachy4.ALL {s.allele) 4 
1 Hairy Nipple Gall (center cell) 
2 Hairy Nipple Gall (side cell) 
3 Glabrous Nipple Gall (center ce 
4 Glabrous Nipple Gall (side cell 
5 Star Gall (center cell) 
6 Blister Gall (center cell) MD 
7 Blister Gall (center cell) TX 
8 Disc Gall (center cell) 
9 Petiole Gall 
1 0 Glabrous Bud Gall 
11 Hairy Bud Gall 
12 Twig Gall 
1 3 Tetrogonocephala 
1-i Celtisaspis 
29 30 
1 1 
1 1 
1 1 
1 1 
1 1 
1 1 
1 1 
1 1 
1 1 
1 1 
1 1 
1 1 
0 0 
0 0 
43 44 
1-PGM J-PGM 
0 0 
1 0 
0 0 
0 0 
0 0 
0 0 
0 0 
0 0 
0 1 
0 1 
0 1 
0 0 
0 0 
0 0 
31 32 33 34 
A-PGM 
0 0 0 0 
0 0 0 0 
0 0 0 0 
0 0 0 0 
0 0 0 0 
0 0 0 0 
0 0 0 0 
0 0 0 0 
1 1 0 0 
1 0 1 0 
1 0 1 0 
1 1 0 0 
. 0 1 
0 0 
45 46 47 48 
K-PGM L-PGM C-PGI E-PGI 
0 0 1 1 
0 0 1 0 
0 0 1 0 
0 0 1 0 
0 0 1 0 
0 0 1 0 
0 0 1 0 
0 0 1 0 
0 0 0 0 
1 0 0 0 
1 0 0 0 
1 0 0 0 
0 0 0 0 
0 1 0 0 
35 36 37 38 
A-PGM B-PGM C-PGM 0-PGM 
0 0 0 1 
0 0 0 0 
0 0 0 1 
0 0 0 0 
0 0 0 1 
0 0 0 1 
0 0 0 1 
0 0 0 1 
0 0 0 0 
0 0 0 0 
0 0 0 0 
0 0 0 0 
1 1 1 0 
0 0 1 0 
49 50 51 52 
F-PGI G-PGI H-PGI 1-PGI 
0 0 0 0 
0 0 0 0 
0 0 0 0 
0 0 0 0 
0 0 0 0 
0 0 0 0 
0 0 0 0 
0 0 0 0 
0 0 0 0 
0 0 0 0 
0 0 0 0 
0 1 1 0 
0 0 0 1 
1 0 0 0 
39 40 
E-PGM F-PGM 
0 1 
0 1 
0 1 
0 1 
0 1 
0 1 
0 1 
0 1 
0 0 
0 0 
0 0 
0 0 
0 0 
0 0 
53 54 
J-PGI K-PGI 
0 0 
0 0 
0 0 
0 0 
0 0 
0 0 
0 0 
0 0 
0 1 
1 0 
1 0 
0 0 
1 0 
0 0 
41 
G-PGM 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
55 
L-PGI 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
42 
H-PGM 
1 
1 
1 
1 
0 
1 
1 
1 · 
0 
0 
0 
0 
0 
0 
56 
A-GPO! 
0 
0 
0 
0 
1 
0 
0 
0 
0 
0 
0 
0 
0 
0 
\ ! 
t ~ 
)---' 
-i::,.. 
\..,.) 
Table 3-2 (cont.) 
Pachy4.ALL (s.a//ele) 5 57 58 59 60 61 62 63 64 65 66 67 68 69 70 
8-GPDt C-GPDt 0-GPDt E-GPDI- F-GPDI- G-GPDI A-ME 8-ME C-ME D-ME E-ME F-ME G-ME H-ME 
1 Hairy Nipple Gall (center cell) 0 1 0 0 0 0 0 0 0 1 0 0 0 0 
2 Hairy Nipple Gall (side cell) 0 1 0 0 0 0 0 0 0 1 1 1 0 0 
3 Glabrous Nipple Gall (center ce 1 1 0 0 0 0 1 0 0 1 0 0 0 0 
4 Glabrous Nipple Gall (side cell 0 1 0 0 0 0 0 0 0 1 0 1 0 0 
5 Star Gall (center cell) 0 1 0 0 0 0 0 0 0 1 0 1 0 0 
6 Blister Gall (center cell) MD 0 1 0 0 0 0 1 0 0 1 0 0 0 0 
7 Blister Gall (center cell) TX 0 1 0 0 0 0 0 0 0 1 0 0 0 0 
8 Disc Gall (center cell) 0 1 0 0 0 0 0 0 0 1 0 0 0 0 
9 Petiole Gall 0 0 0 0 1 0 1 0 0 0 0 0 0 0 
10 1 Glabrous Bud Gall 0 0 0 0 0 1 0 0 0 0 0 0 1 0 
11 \ Hairy Bud Gall 0 0 0 0 0 1 0 0 0 0 0 0 1 1 
1 2 \ Twig Gall 0 0 1 0 0 0 0 0 0 0 0 0 0 1 
1 3 \ Tetrogonocephala 0 I 1 0 I 0 0 0 0 1 0 0 0 0 0 0 
1 4 \ Celtisaspis \ 0 I 1 I 0 I 0 \ 0 \ 0 0 I 0 1 I 0 I 0 I 0 0 I 0 
Pachy4.ALL (s.al\e\e) 6 1 I 72 I 73 1 74 i 75 I I 83 84 71 76 77 78 79 80 81 82 
I-ME J-M E K-ME A-FUM B-FUM C-FUM D-FUM E-FUM F-FUM G-FUM H-FUM B-PEP· C-PEP· E-PEP1 
1 \ Hairy Nipple Gall (center cell) 0 0 0 0 0 0 1 0 0 0 0 1 1 0 
2 \ Hairy Nipple Gall (side cell) 
' 
0 \ 0 I 0 I 0 I 0 0 I 1 I 0 I 1 0 0 1 1 1 
3 \ Glabrous Nipple Gall (center ce 1 0 I 0 \ 0 I 0 0 0 1 0 0 0 0 1 1 1 
4 Glabrous Nipple Gall (side cell 0 0 0 0 0 0 1 0 0 0 0 1 1 0 
5 Star Gall (center cell) 0 0 0 0 0 0 1 0 1 0 0 1 1 0 
6 Blister Gall (center cell) MD 0 0 0 0 0 0 1 0 1 0 0 0 1 0 
7 Blister Gall (center cell) TX 0 0 0 0 0 0 1 0 1 0 0 0 1 0 
8 Disc Gall (center cell) 0 0 0 0 0 0 1 0 0 0 0 1 1 0 
9 Petiole Gall 0 0 0 0 0 0 1 1 1 1 0 1 1 0 
10 Glabrous Bud Gall 1 1 1 0 0 1 0 0 0 0 0 0 0 1 
11 Hairy Bud Gall 1 1 1 0 0 1 0 0 0 0 0 0 0 1 
12 Twig Gall 0 0 0 1 0 1 0 0 0 0 0 0 0 0 
13 Tetrogonocephala 0 0 0 0 0 0 1 0 1 0 0 0 0 0 
14 Celtisaspis 0 0 0 0 0 0 0 0 1 0 0 0 0 0 
~ 
........ 
~ 
~ 
Table 3-2 (cont.) 
Pachy4.ALL (s.allele) 7 85 86 87 88 89 90 91 92 93 94 95 96 97 98 
F-PEP1 G-PEP· H-PEP· I-PEP1 A-FOP B-FDP B-TPI C-TPI 0-TPI E-TPI A-G3Df B-G3Df A-IDH1 B-IDH1 
1 Hairy Nipple Gall (center cell) 0 0 0 0 0 1 1 0 0 0 0 1 0 0 
2 Hairy Nipple Gall (side cell) 0 0 0 0 0 1 1 0 0 0 0 1 0 0 
3 Glabrous Nipple Gall (center ce 0 0 0 0 0 1 1 0 0 0 0 1 0 0 
4 Glabrous Nipple Gall (side cell 0 0 0 0 0 1 1 0 0 0 0 1 0 0 
5 Star Gall (center cell) 0 0 0 0 0 1 1 0 0 0 0 1 0 0 
6 Blister Gall (center cell) MD 0 0 0 0 0 1 1 0 0 0 0 1 0 0 
7 Blister Gall (center cell) TX 0 0 0 0 0 1 1 0 0 0 0 1 0 0 
8 Disc Gall (center eel\) 1 0 0 0 0 1 1 0 0 0 0 1 0 0 
9 Petiole Gal\ 0 0 0 0 0 1 0 0 1 0 0 1 0 0 
10 Glabrous Bud Gall 0 0 0 0 0 1 1 0 0 1 0 1 0 0 
11 Hairy Bud Gal\ 0 0 0 0 0 1 0 0 0 1 0 1 0 0 
12 Twig Gall 1 0 0 0 0 1 0 1 0 0 0 1 0 0 
13 Tetrogonocephala 0 0 0 1 0 1 1 0 0 0 0 1 0 0 
1 4 \ Celtisaspis I 0 I 1 I 1 I 0 1 0 I 1 0 0 I 0 I 1 0 1 1 
Pachy4.ALL (s.a\\ele) 8 1 99 1 100 I I 107 I 101 102 103 104 105 106 108 1 09 110 111 112 
C-IDH1 D -lDH1 E -lDH1 F -lDH1 A-IDH, B -IDH2 D -IDH2 G-IDH2 A-MOH B-MDH D-MDH E-MDH F-MDH G-MDH 
1 I Hairy Nipple Gall (center eel\) 0 1 1 0 0 0 1 0 0 1 1 0 0 0 
2 Hairy Nipple Gal\ (side eel\) 0 1 1 0 0 0 1 0 0 0 1 1 1 0 
3 Glabrous Nipple Ga\\ (center ce 0 1 1 0 0 0 1 0 1 1 1 0 0 0 
4 Glabrous Nipple Ga\\ (side eel\ 0 0 1 0 0 0 1 0 0 1 1 1 1 1 
5 Star Ga\\ (center eel\) 1 1 1 1 0 0 1 0 1 1 1 0 0 0 
6 Blister Ga\\ (center eel\) MD 0 1 0 0 0 0 1 0 0 1 1 0 0 0 
7 Blister Gal\ (center eel\) TX 0 1 0 0 0 0 1 0 0 1 1 0 1 0 
8 Disc Gal\ (center eel\) 0 1 1 1 0 0 1 0 0 1 1 0 1 0 
9 Petiole Gall 0 0 0 0 0 0 0 1 0 0 1 0 0 0 
10 Glabrous Bud Gal\ 0 0 1 0 0 0 0 1 0 1 1 0 0 0 
11 Hairy Bud Gal\ 0 0 1 0 0 0 0 1 0 0 1 0 0 0 
12 Twig Gall 1 0 0 1 0 0 0 1 0 0 1 0 0 0 
13 Tetrogonocephala 0 0 0 1 0 1 0 0 0 0 0 0 0 1 
14 Celtisaspis 0 0 0 0 1 0 0 0 0 0 0 0 0 1 
,_. 
.i:,.. 
\J\ 
Table 3-2 (cont.) 
Pachy4.ALL (s.alle/e) g 113 114 115 116 117 11 8 119 120 121 122 
H·MOH /-MOH · A-MOH 8-MOH A·6PGI 8-6PG! C-6PG! 0-6PG! E-6PG[ F·6PGC 
1 Hairy Nipple Gall (center cell) 0 0 0 1 0 0 0 1 0 1 
2 Hairy Nipple Gall (side cell) 0 0 0 1 0 0 0 1 0 0 
3 Glabrous Nipple Gall (center ce 0 0 0 1 0 0 0 1 0 0 
4 Glabrous Nipple Gall (side cell 0 0 0 1 0 0 0 1 1 1 
5 Star Gall (center cell) 0 0 0 1 0 0 0 1 0 1 
6 Blister Gal\ (center eel\) MD 0 0 0 1 0 0 0 1 0 1 
7 Blister Gal\ (center cell) TX 0 0 0 1 0 0 0 1 0 1 
8 Disc Gal\ (center eel\) 0 0 0 1 0 0 0 1 0 0 
9 Petiole Gal\ 0 1 0 1 1 1 0 1 0 0 
1 O \ Gtabrous Buel Gall 0 0 0 1 0 0 0 0 0 0 
11 \ \--\airy Bud Gall 0 0 0 1 0 0 0 0 0 0 
1 2 \ "Twig Gall 1 0 0 1 0 0 1 0 0 0 
13 \ 1etrogonocephala 0 0 1 0 1 1 1 0 0 0 
1 4 \ Celtisaspis \ 0 I 0 I 1 I 0 I 0 0 I 0 I 0 I 0 I 0 
?achy4.ALL ts.allele) 10 I 121 I 12s I 12s I 130 I 131 I 132 I 133 I 134 I 13s 
C-LD\--1 \D-LD\--1 \E-LD\--1 \F-LD\--1 \m-kan\f-karv,\voltine o verwi ovioosi 
1 \ \--\airy Nipple Ga\\ (center ce\\) \ 0 I 1 \ 1 \ 0 I 5 I 6 \ 1 2 1 
2 I \--\airy Nipple Ga\\ (side ce\\) 
' 
1 I 1 0 0 \ 5 6 1 2 1 
3 G\abrous Nipple Ga\\ (center ce 0 1 1 0 5 6 1 2 1 
4 G\abrous Nipple Ga\\ (side eel\ 1 1 0 0 ? ? 1 2 1 
5 Star Ga\\ (center eel\) 0 1 0 1 ? ? 1 2 1 
6 Blister Ga\\ (center cell) MD 1 1 0 1 5 6 1 2 1 
7 Blister Gall (center eel\) "TX 0 1 1 1 ? ? 1 2 1 
8 Disc Gal\ (center cell) 0 1 0 0 ? ? 1 2 1 
9 Petiole Ga\\ 0 0 0 0 3 4 1 1 0 
10 G\abrous Bud Gall 0 1 0 0 2 2 1 1 2 
11 Hairy Bud Ga\\ 0 1 0 0 ? ? 1 1 2 
12 Twig Gall 0 1 0 0 5 6 1 1 2 
13 1 etrogonocephala 0 1 0 1 5 6 2 ? ? 
1 4 Celtisaspis 0 0 0 0 ? ? 1 / 2 0 ? 
123 124 
G-6PG! H·6PGC 
1 0 
0 0 
0 0 
0 0 
0 0 
0 0 
0 0 
1 1 
0 0 
0 1 
0 1 
0 0 
0 0 
I 1 0 
125 
A-LOH 
0 
0 
0 
0 
0 
0 
0 
0 
1 
0 
0 
0 
0 
1 
126 
8-LOH 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
0 
1 '\ i. 
\ 
\ 
~ 
\ 
Tab1e3_3 L' 
• 
1
st of morphological characters used for Pachypsylla phylogenetic study 
I. ADULT 
l-IEAD 
1. Surface t . 
0 exture of vertex [seen from SEM (Fig. 3-8)]: 
· rough surface without obvious pattern, with scattered hairs, e.g. 
C~!tisaspis, 
1. With small shallow bumps like fish scales, with scattered hairs, e.g. 
T~tragonocephala flava, 2
· With shallow wrinkles and some scattered short setae. e.g. Pachypsylla 
bud galler, 3
· densely and evenly covered with deep wrinkles and short setae, e.g. 
Pachypsylla leaf and petiole gallers. 
2
' ~etween vertex and gena [seen from dry specimens]: 
· gena conical, not forming a flat plane vertical to body, gap between 
frons and vertex not deep, . 
1. frontal part of gena very flat, vertical to the body, formrng a grove 
between vertex and gena. 
3
· gurfac~ texture of genae [seen from SEM] (Fig. 3.-8): 
l.· a little r.ough, e.g. Tetragonocephala, Celtisas_p_zs, 
2 smooth m frontal portion and rough on the ~1de, e.g. ~ud gaJJer, 
· bumpy, with deep wrinkles, e.g. leaf and petwle gallers. 
4. L 
0 ength of setae on genae: 
1 · ~e~rly homogeneous, . . . 
. apical end of genae with group of s1gmficantly longer setae. 
5. Le 
0 ngth of genae + length of vertex: 
· < 1/2 
6. 
7. 
1. ~ 1 
OAntenna length + head width: 
· > 1.2 
1. ~ 1.0 
Thick · 1 
0 ness of antenna, distal vs. prox.1ma 'th 'milar width between 
· flagellum of antenna within each segment wi si 
distal and proximal end.s, . 1 end and wider distal end, l. flagellum of antenna with narrow proxima 
esp. seg. VI to VIII. 
146 
Tl-IoRAx 
8. Surface f h . 
0 ° t orac1c dorsum: 
· cover d · h 
1 1 
e Wit conspicuous, short stiff pubescence 
· g abro h. · ' us, s mmg, often with sparse minute pubescence. 
9. Pterostigma 
O. indistinct 
1. distinct. ' 
10. Bind-cox 
11. 
12. 
out . ae, ventral crescentic denticulate patch of nearly crescent shape on 
0 er _side above meracanthas [seen under SEM (Fig. 3-ll)]: 
1. With some bumps, not pointed 
. full . ' With sharply pointed bird-beak-structure. 
Fan-shaped . . (F' area m crescent-shaped patch on hmd-coxa, [seen under SEM 
ig. 3-ll)]: 
O. no ob · 
1 . v10us groves and ridges. 
2 · w~ th a few ridges parallel to each other, 
· With obvious groves and ridges in fan-like arrangement. 
teracanthus, length + width: 
· > 2.0 
I. l.O - 1.5 
13. T 
0 wo black claws on the basal segment of metatarsus 
· absent 
1
· presen~. 
~LE GENITALIA 
· Male · · 1 1 · 
0 Proct1ger bipartite, setae on the basal segment m at:ra view: 
· sparsely distributed on apical caudal area, e.g. Leaf bhster galler, and 
outgroups 
1. apical caudal area and caudal margin densely covered. 
15
· Setae · · 
0 ' ~p1cal half of basal segment of male proct1g~r: 
· m_amly on upper caudal side, leaving sma11 port10n of lower frontal part 
Without setae 
l. evenly distrib~ted. 
16. 
~ale forceps, in lateral view: . , , 
· gradually tapering at apically, lateral margms convex, not parallel. 
1. abruptly reduced subapically, lateral margins parallel. 
147 
I 
I 
I ( 
I 
17. i\ -r 
lVlale aedea 
0. round dgus, apex of head of the second segment: 
1 h e , blunt. 
· ooked. 
18. 
Second se . 
0. formi;ment of male aedeagus, base of head at joint to basal rod: 
1. · . . g a sharp angle 
Joirung smoothly. ' 
F'Ei\fA 
19. p L~ GENITALIA 
aired thumb 1 'k 
i-la): - 1 e processes, lateral proximal side of female proctiger (Fig. 
1 · absent, 
· Present. 
20. 
Shape of£ 
0. Wid emale dorsal valve (proctiger): 
1. sho~ :t base, pointed toward the end, triangular shape in lateral vi.ew, 
2l. ' ounded down, broad at caudal end, pointed toward ventral side. 
Setae on£ 
0. s emale proctiger 1 etae of h . 
. long th. omogeneous size evenly distributed, 
Tetro 
O
m setae densely covering apex, shorter setae elsewhere, e.g. 
23. '!} nocepha!a, Ce!tisaspis. 
Peznale circ 
0. sing) umanal ring, rows of pores 
1 e row 
· do b ' u le row lI · 
·~PfI( 5th instar) 22 
. Dorsal su 
0 1 rface of head· 
· acki . · 
1. 'With ng _consp1cuous chitinized areas. 
24. Pa
1red rectangular chitinized bands. 
Base of n 
0. ve Ymphal antenna· . 
s ry close to margi~ of head the antenna extending beyond first 
1. n:grnent outside margin of head. · 'd 
ar mar . d f econd segment rns1 e the h gm, the entire first segment an part o s 
2. f' ead margin 
irst and ' . 25. second segments inside margm . 
l'roch o anter: 
1 · absent, 
· Present. 
148 
26. Pairs of ventral plates on abdomen anterior to caudal plate: 
0. no plates, 
1. only one pair near the caudal plate, 
2. three pairs, each on a different segment. 
27. Abdominal median dorsal plate near caudal plate: 
0. absent, 
1. present, one small irregular plate, e.g. twig galler, 
2. paired small semi-circular plates, e.g. most leaf gallers. 
28. Dorsal margin of caudal plate in conjunction with abdomen: 
0. a straight line without median indentation, 
1. a straight line with median indentation, i.e. a small middle area on the 
margin not chitinized. 
29. Caudal plate in ventral view: 
0. a median plate flanked by pair of ventral plates, 
1. entire, except for segmentation, without separate ventra l plate. 
30. Pointed chitinized caudal spurs: 
0. absent, 
1. present. 
31. Central apical spur on pointed chitinized caudal spurs (if present): 
0. sharp pointed, 
1. blunt and/or notched. 
32. Apical caudal spurs on strongly chitinized caudal plates: 
0. present in apical two segments only, 
1. present in alJ three segments. 
33. Abdominal caudal spurs: 
0. absent, 
1. present. 
149 
~~~I~ 4-l . Classification of Spondyliaspidinae, showing differences between systems of White & 
indicatson and Burckhardt. Asterisks indicate taxa included in my phylogenetic analyses. Underlining 
es taxa whose position has been especially controversial. 
~ITE & 
SPONDy HODKINSON (1985) 
Are _!-,lASPIDIDAE 
PUunae White & Hodkinson 
~* 
Euphaler · B E mae ecker-Migdisova 
~* Retroac · · H . ? izzia eslop-Harnson 
.Pacfzyparia Longinova 
~* 
Co!p!zorina Capener* 
-* 
Pachyp Jr p sy inae Becker-Migdisov 
achypsy/la Riley* 
Tetrag 
onocephala Crawford* 
PachJps 11 . 
. Y a ;aponica Miya take * 
Spondy!" · · iaspidmae Schwartz 
A ustra/opsy/la Tuthill & Taylor 
Cardiaspina Carwford* 
Creiis Scott* 
Eucalyptolyma Froggatt 
G!"ca · 
., sp1s Taylor* 
Hya/inaspis Taylor 
Lasiopsy/la Froggatt 
Spolldyfiaspis Signoret* 
APt.r , 
' V\LA RIDA E 
Ctenaryt · · 
a znmae White & Hodkinson 
er* 
Syl!ca · 
E 
1P1olyma Froggatt 
Url,i l lloco a Crawford 
BURCK.HARDT (1991) 
=Russelliana* (APHALAROIDINAE) 
( out of Spondyliaspidinae) 
( If ti " ) 
( " " ) 
( = Phyllolyma Walker) 
SPONDYLIASPIDINAE 
Pachypsyllini 
Pachypsylla* 
Tetragonocepha!a * 
Celtisaspis Yang & Li* 
Spondyliaspidini Schwartz 
Australopsylla 
Boreiogycaspis Moore 
Cardiaspina* 
Creiis* 
Dasypsylla Froggatt 
Eucalyptolyma Froggatt 
Eurhinocola Crawford 
Glycaspis Taylor* 
Hyalinaspis Taylor 
Kenmooreana Taylor 
Lasiopsylla Froggatt 
Phellopsy!!a Taylor* 
Phy!!olvma Walker* 
Platyobria Taylor 
Spondyliaspis Signoret* 
(to Spondyliaspidini Schwartz), including: 
Agelaeopsylla Taylor 
Anoeconeossa Taylor 
Blastopsylla Taylor 
C,yptoneossa Tay!or * 
Ctena,ytaina Ferns & Klvver 
Eriopsylla Frogga tt 
Leptospermonastes Taylor 
Syncarpiolyma Frogga tt 
150 
Table 4-2 L' 
Phylog · . 1st of characters used for Pacbypsyllini/Spondyliaspidinae 
enet1c study. 
t~DLT CHARACTERS. 
1. Vertex· th 1 . 
0 · e owest part of the lower margin, adJacent to gena: 
· near median ocellus 
1 ' 
· near antennal socket 
2 ' 
3 · betw~en ocellus and antennal socket, 
· margm flat, equally low. 
2
· Angle between top margin of vertex and median carina: 
0. ~ 900 
1. > > 900. 
3. lrregul d . 
ar epress10ns or fovea on vertex: 
O. not obvious i' one obvious' depression, 
3 · two apparent depressions, 
· three apparent depression. 
4. Genal cones: 
0. absent. 
i_' present, length < < 1/2 length of vertex, 
present, length about half as long as vertex, 3
· present, subequal to or longer than vertex in length. 
5. Shape of occipital foramen surrounded by the occipital sclerite, when 
Vertex · · ' pos1t10ned horizontally: 
O. basic shape dorsal margin straight (Fig. 4-2 a), 
1. similar to a' dorsal margin indented (Fig. 4-2 b ), . 
2
· reverse tria~gular shape, ventral margin poninted (Fig. 4-2 c). 
6. l 
ength of antennae ..,.. width of head including eyes: 
O. l.71-20 
1. 1.11-1.69,' 
2. 1 ~ .0, 
3
. < 0.7. 
7. 
AntennaI segment X length + segment I length: 
0. s 0.6 
1. 0.61 ~ 1 O 
2. > 1.0 .. ' 
151 
S. Antennal segment width, distal to proximal: 
1. Widen at distal end within each segment, especially on seg. VI to 0. homogeneous, 
VIII. 
9
· Rhinaria: 
0. absent from seg. V, 
1. present on seg. V. 
10. An teoccipital lobes, a usually narrow sclerite between eye and 
genae /vertex: 
0. present, obvious, 
1. highly reduced or absent. 
TBORAX 
11 Ed 
· ge of forewing apex: 
0. rounded 
1. pointed.' 
12
· Forewing: 
0. not elongate with curvy veins not parallel, 
I. very elongate with veins straight and almost parallel: 
13. Pterostigma: 
O. present, obvious, 
1. absent or very reduced. 
14. Number of spines, thick black saltatorial spurs, at apex of metatibia: 
0. < 6 
' 1. ~ 6. 
15. M etatibial basal (genual) spine: 
0. present 
1. absent.' 
16. A pical spines on metatarsal segment I: 
0. present, 
1. absent. 
17. M eracanthus: 
0. prominent, 
1. absent. 
152 
MALEG 18. Mal ENITALIA 
e proctiger: 
1 
b' ai:tite (seg. X and XI of abdomen fused) 0. unip · 
· 1part1te. ' 
19. Male par 
0 
_amere, lateral view: 
1 · d~stal end tapered, 
. distal end swollen. 
20. M ale para 
0 
mere, lateral protruding lobes: 
· absent 
1. present~ 
21. M ale para . 
0 
mere, a !me of marginal teeth at terminal end: 
· absent 
1. presen~. 
22. Shape of . o. 
1 
ternunal segment of aedeagus, lateral view: 
ong and narrow with a small bead, like a slender rod swollon at 
one end 
1 ' 
2
· ~wollen before middle without a slender rod-like part at the base, 
FE: · ramatically swollen at both ends. 
23~:,E GENITALIA 
ape off 1 
0 
. ema e dorsal valve (proctiger). lateral view: 
1
· wide at base, pointed toward the end, tiangu!ar, 
· wide at base till middle, at apical 1/3 narrow, sharply angled 
dorsally, 
2
· short, rounded down broad at caudal end, pointed toward vetral 
side. ' 
24. S etae on f o . ~male proctiger: 
1
: surular i~ size, evenly distributed, . uneven m size, a group of Jong setae ctorsomed1anly, others shorter 
and evenly distributed in rest area, 
2
· long thin setae densely covering the apex, some short setae 
dorsally. 
25. N umber of . I . o . rows of pores on female circumana area. 
. smgle 
1. doubl
1 
e. 
153 
26. Shape of f . 
0 
.e~ale circumanal ring 
· elliptical/ circular 
1. convoluted. ' 
~~· NYMPHS 
t-iEAD 
27. Number f 
0 
° nymphal antenna! segments: 
· 10 
' 1. 9. 
28. Base of 
· very close to bead margin, only partia!l first basal segment cover by 
0 
nymphal antenna placement: 
1. 
2. 
head 
' partial second segment covered by head, 
both the first and second segments covered by head. 
TBORAX 
29 A . 
· p1cal 
0 
an~le of nymphal forewing-pad: . . 
1
· ~dJa~ent or exterior to the margin of the hmdwmg-pad, 
· mtenor to the hindwing-pad margin, 
30. Tr h oc anter: 
31. 
O. absent 
1. presen~. 
Arolium· 
0. 
1. 
triangular, 
32. N 
2. 
broad, semi-circular 
highly reduced or absent. 
Ymphal unguitractor, a central sclerotized rod within arolium: 
~- present, obvious, 
· absent or very reduced. 
~DO MEN 
. Ny mphal abdominal sclerites on dorsal surface: 
~- p~esent, obvious, . . . . 
· highly reduced or lacking distinct sclente. 
34. Nym h 
0 
P al abdominal apex: 
1 
· sm.oothly truncate, 
· pomted. 
154 
35. Sh a{{° of nymphal abdominal apex, if pointed: 
1 
· no special structure, 
· serrate 
2 . ' 
· with large medial 'teeth'. 
36. N 
P
{':'Ph_al abdominal segments produced laterally as rounded or 'tooth' like 
OJect1ons: 
0. absent 
1. presen~. 
37. La teral bulges on the abdomen, coincident with each apparent segment: 
0. absent 
1. prese;t. 
38
· Position f . o anal operung: 
0. ventral 
1. most p~sterior point. 
39. Circum 1 . 
-ana pore nng (nymphs): 
O. present 
1. absent.' 
40. An~ pore field ( other than circum-anal ring): 
· absent 
1. presen't. 
4
1. Ar 1 · 'f t 
r~ngement of anal pores in field other than circum-ana nng, 
1 
presen : 
· entire or broken rings 
1 ' 
· small groups. 
42. Setal t 
0 
YP.e on nymphal head: 
1 
· simple setae only, 
· clavate setae 
2 . ' 
· cap1tate setae. 
43. S etal type, beside simple setae on dorsal area of nymphal abdomen: 
0 . ' 
1 
· simple setae only, 
· clavate setae 2 . ' 
· cap1tate setae. 
155 
44. S 
etaI tYPe, beside simple setae, on ventral area of nymphal abdomen: 
0. simple setae only, 
1. lanceolate setae 
2. rod setae ' 
3. capitate s~tae. 
45. s 
etaI ~e, beside simple setae, on the margin of nymphal abdomen: 
O. simple setae only, 
l. Ianceolate setae 
2. ' secta setae 
3
. capitate se;ae. 
156 
,-..,. 
VI 
-..l 
Table 4-3. Character matrix used in Spondyliaspidinae phylogenetic analyses based on morphological 
characters. 
spon4.data 1 1 2 3 4 5 6 I 7 8 9 1 0 11 12 13 14 
ve rtex ve r tex fovea Qenalc occipit, A L/HWI XII ant. rhinaric an teocc w inQ a1 w inQsh, pt erost s altato1 
1 Pachypsyl/a c-mam ma 2 0 1 2 1 2 0 1 0 1 0 0 0 1 
2 Pachypsylla c -vesicu/a 2 0 1 2 1 2 0 1 0 1 0 0 0 1 
3 Pachypsylla venusta 2 0 1 2 1 2 0 1 0 1 0 0 0 1 
4 Pachypsylla c-gemma 2 0 1 2 0 2 1 1 0 1 0 0 0 1 
5 Tetragonocephala flava 3 0 1 2 0 1 1 0 0 1 0 0 0 1 
6 Celtisaspis bei jingana 3 0 1 3 0 1 0 0 0 1 0 0 0 1 
7 Phellopsylla spp . 1 0 3 1 1 1 0 0 0 0 1 0 0 1 
8 Phyllolyma rufa 1 0 3 1 1 3 0 1 0 0 1 0 0 0 
9 Phyllolyma sp 1 0 3 1 1 3 1 1 0 0 1 0 1 0 
1 o I Creiis tecta 2 0 1 2 1 0 1 0 0 0 0 1 0 1 
11 \ Creiis \ongipennis 2 I 0 1 2 1 0 1 0 0 0 0 1 0 1 
1 2 \ Spondyliaspis p\icatu\oid 2 I 0 \ 1 2 1 1 2 0 0 0 0 1 0 1 
1 3 \ Cardiaspina a\bitexturata 3 I 0 I 1 2 1 2 1 0 0 0 0 1 0 1 
1 4 \ Cardiaspina vittaformis 3 I 0 I 1 I 2 1 2 1 0 0 0 0 1 0 1 
1 5 \ Glycaspis bai\eyi 2 I 0 I 1 I 3 2 0 1 0 1 0 1 1 0 0 
1 6 \ G\ycaspis aggregata 2 I 0 I 1 I 3 I 2 0 1 0 1 0 1 1 0 0 
1 7 \ Glycaspis plani\ecta 2 I 0 I 1 I 3 I 2 0 1 0 1 0 1 1 0 0 
1 8 \ Ctenarytaina eucalypti 2 I 0 I 1 I 2 I 1 2 2 1 0 0 0 1 0 0 
1 9 \ Co\ophorina spn 0 I 0 I 0 I 2 I 0 I 2 2 0 0 1 0 0 0 0 
2 O \ Euphalerus spA 0 I 0 I 1 I 2 I 0 I 1 2 0 0 1 0 0 0 0 
2 1 \ Euphalerus nidifix 0 I 0 I 1 I 2 2 I 1 2 0 0 0 0 0 0 0 
2 2 \ Russe\\iana adesmiae 0 I 1 0 2 0 1 1 0 0 1 0 0 0 0 
2 3 \ Russe\\iana fabiahae 0 1 0 2 0 1 1 0 0 1 0 0 0 1 
2 4 \ Heteropsyl\a texana 0 0 0 0 2 0 2 0 0 1 0 0 0 0 
2 5 Acizzia uncatoides 0 1 2 2 2 1 0 0 0 1 0 0 0 0 
26 Trigonon longicomis 0 1 I 2 0 2 0 2 0 0 1 0 0 0 0 
27 1 Psylla a\ni 
' 
0 I 1 I 1 I 2 2 I 0 I 1 0 I 0 1 0 I 0 0 0 
..... ~~~~~~~ 
,-,. 
\JI 
00 
Table 4-3 (cont.). 
spon4.data 2 15 1 6 17 18 19 20 21 22 23 24 25 26 27 28 
oen ual metata1 me race male c pa ram1 pa ram1,pa ram1 ae deaq, le male se tae circum, cir cum, an t sea an t ola 
1 Pachypsylla c-mamma 1 0 0 1 0 0 0 0 0 0 0 0 0 0 
2 Pachypsylla c -vesicula 1 0 0 1 0 0 0 0 0 0 0 0 0 0 
3 Pachypsylla venusta 1 0 0 1 0 0 0 0 0 0 0 0 0 1 
4 Pachypsylla c-gemma 1 0 0 1 0 0 0 0 0 0 0 0 0 1 
5 Tetragonocephala !lava 1 1 0 1 0 0 0 0 1 2 1 0 0 0 
6 Celtisaspis beijingana 1 0 0 1 0 0 0 0 1 2 1 0 0 2 
7 Phe\\opsy\\a spp. 1 0 1 0 0 0 0 0 0 0 0 0 1 0 
8 \ Phy\\olyrna ruta 1 0 1 0 0 0 0 ? ? ? ? ? ? ? 
9 I Phy\\olyrna sp 1 I 0 \ ? 0 0 0 0 0 ? ? ? ? 1 0 
1 o \ Creiis tecta 0 I 0 I 1 1 1 0 1 1 ? ? ? ? ? ? 
11 \ Creiis \ongipennis 0 I 0 I 1 I 1 1 0 1 1 0 1 1 0 0 0 
1 2 I Spondy\iaspis p\icatuloid 0 I 0 I 1 I 1 ? ? ? ? 0 1 1 1 0 0 
1 3 \ Cardiaspina albitexturata 0 I 0 I 1 I 1 1 0 1 0 0 0 1 0 0 0 
1 4 \ Cardiaspina vittaforrnis 0 I 0 I 1 1 1 0 1 0 ? ? ? ? 0 0 
1 s I Glycaspis bai\eyi 0 I 0 I 1 1 1 1 0 1 0 0 1 2 0 0 
1 6 I G\ycaspis aggregata 0 I 0 I 1 1 1 1 0 1 0 0 1 2 ? ? 
1 7 \ G\ycaspis planitecta 0 I 0 I 1 1 1 0 0 1 ? ? ? ? ? ? 
1 8 \ Ctenarytaina eucalypti 1 1 I 0 1 0 0 0 0 0 0 1 0 0 0 
1 9 \ Colophorina spn 1 0 0 0 0 0 0 0 ? ? ? ? 1 0 
2 O I Eupha\erus spA 1 I 0 0 0 0 0 0 0 0 1 1 0 1 1 
2 1 I Eupha\erus niditix 0 I 0 0 0 0 0 0 0 0 1 1 0 1 2 
22 I Russelliana adesrniae 0 1 0 0 1 0 0 1 0 1 0 0 ? ? 
23 Russelliana tabiahae 0 I 1 0 0 1 0 0 1 0 1 0 0 ? ? 
24 Heteropsy\\a texana 0 I 0 0 0 0 1 0 2 0 0 1 0 1 0 
25 Acizzia uncatoides 0 
. 
0 0 0 0 1 0 2 0 1 1 0 1 1 
26 Trigonon \ongicornis 0 0 0 0 0 0 0 ? 0 1 1 0 ? ? 
27 \ Psy\\a a\ni I 0 I 0 \ 0 0 1 \ I 0 I 1 I 0 0 1 1 0 1 1 
r-' 
Ul 
\Q 
Table 4-3 (cont.) . 
spon4.data 3 
1 Pachypsylla c-mamma 
2 Pachypsylla c -vesicula 
3 Pachypsylla venusta 
4 Pachypsylla c-gemma 
5 Tetragonocephala !lava 
6 Celtisaspis beijingana 
7 Phel\opsylla spp . 
8 Phyllolym a rufa 
9 Phyllolyma sp 
10 Creiis tecta 
11 Creiis longipennis 
12 Spondyliaspis plicatuloid 
13 Cardiaspina albitexturata 
14 Cardiasp ina vittaformis 
1 5 Glycaspis bai\eyi 
1 6 Glycaspis aggregata 
17 Glycaspis plani tecta 
18 Ctenaryta ina eucalypti 
19 Colophorina spn 
20 Euphalerus spA 
21 Euphalerus nidifix 
22 Russelliana adesmiae 
23 Russell iana fabiahae 
24 Heteropsylla texana 
25 Acizzia uncatoides 
26 Trigonon longicornis 
27 Psylla alni 
29 I 30 
win-pa<I trocha1 
0 1 
0 1 
0 1 
0 1 
0 1 
0 1 
0 0 
? ? 
0 0 
? ? 
1 0 
0 0 
1 0 
1 0 
0 0 
? ? 
? ? 
0 0 
0 0 
0 0 
0 1 
? ? 
? ? 
0 0 
0 0 
? ? 
0 0 
31 32 I 33 
aroliun unQuit d dorsal 
2 1 I 1 
2 1 I 1 
2 1 I 1 
2 1 1 
2 1 1 
2 1 1 
2 1 0 
? ? ? 
2 1 0 
? ? ? 
2 1 0 
2 1 0 
2 1 0 
2 1 0 
2 1 0 
? ? ? 
? ? ? 
2 1 0 
0 0 0 
0 0 0 
0 0 0 
? ? ? 
? ? ? 
1 0 0 
1 0 0 
? ? ? 
1 0 0 
34 I 35 36 37 38 39 40 41 42 
abd spi pointed lat. prc abd lat anal op circum- anal pc anal pc head s, 
1 I 2 0 0 1 1 1 1 0 
1 2 0 0 1 1 1 1 0 
1 2 0 0 1 1 1 1 0 
1 2 0 0 I 1 1 1 1 0 
1 0 0 0 1 1 1 1 0 
1 0 0 0 1 1 1 1 0 
0 0 0 1 0 1 ? 0 
? ? ? ? ? ? ? ? ? 
1 0 0 0 1 0 1 1 0 
? ? ? ? ? ? ? ? ? 
1 0 1 1 1 1 1 1 0 
-· 
1 0 0 0 1 1 0 0 
1 0 0 1 1 1 0 0 
1 0 0 1 I 0 1 0 0 
1 0 0 0 / 1 1 1 1 2 0 
? ? ? ? ? ? ? ? ? 
? ? ? ? ? ? ? ? ? 
0 0 0 1 0 1 1 0 
0 - 0 0 1 0 1 0 0 
0 - 0 0 1 0 1 0 0 
1 1 0 0 1 1 1 0 0 
? ? ? ? ? ? ? ? ? 
? ? ? ? ? ? ? ? ? 
0 0 0 0 0 0 - 1 
0 0 0 0 0 0 2 
? ? ? ? ? ? ? ? ? 
0 0 0 1 0 0 0 
l 
,..--
....... 
°' 0 
Table 4-3 (cont.). 
spon4.data 4 
1 Pachypsylla c-mamma 
2 Pachypsylla c -vesicula 
3 Pachypsylla venusta 
4 Pachypsyl!a c-gemma 
5 Tetragonocephala !lava 
6 Celtisaspis beijingana 
7 Phe\lopsylla spp . 
8 Phyllolyma rula 
9 Phy\lolyma sp 
10 Creiis tecta 
11 Creiis longipennis 
1 2 I Spondyliaspis plicatu\oid 
1 3 \ Cardiaspina albitexturata 
1 4 \ Cardiaspina vittatormis 
1 s \ Glycaspis baileyi 
1 6 I G\ycaspis aggrega\a 
1 7 I Glycaspis planitecta 
1 8 \ C\enary\aina eucalypti 
1 9 \ Colophorina spn 
2 o \ Euphalerus spA 
2 1 \ Euphalerus niditix 
2 2 \ Russelliana adesmiae 
2 3 \ Russe Iliana tabiahae 
2 4 \ Heteropsylla texana 
2 s \ Acizzia unca\oides 
26 \ Trigonon longicomis 
2 7 \ Psylla alni 
43 44 45 
ab d se1 abd se abd sel 
0 0 0 
0 0 0 
0 0 0 
0 0 0 
0 0 0 
0 0 0 
0 1 1 
? ? ? 
0 1 1 
? ? ? 
0 0 0 
0 I 0 0 
0 I 0 \ 0 
0 I 0 I 0 
0 \ 0 \ 0 
? \ ? \ ? 
? I ? I ? 
0 I 0 I 1 
0 I 0 I 2 
0 I 0 I 2 
0 I 0 I 0 
? I ? I ? 
? I ? I ? 
1 I 2 I 1 
2 3 3 
? ? ? 
I 0 I 0 I 0 
........ 
°' ........ 
Table 4-4. Summary of concealment types and character states scored for each taxa for the analyses of evolu tion of 
gaII / lerp formation of Spondyliaspidinae. 
Taxa Coding! Coding2 Coding3 Summarv of concealed biologv** 
Pachypsylla celtidismamma 2 2 0 G 
Pachypsylla celtidivesicu la 2 2 0 G 
Pachypsylla venusta 2 2 0 G 
pachypsylla celtidisgemma 2 2 0 G 
Tetragonocephala flava 1 0 2 L 
Celtisaspis beijingana 1/ 2 1 2 L+G(P) or L 
Phellopsylla spp. 0 0 1 F (W) [under bark w/ floccuent cover] 
Phyllolyma rufa 1 0 2 L 
Phyllolyma sp. 1 0 2 L 
Creiis tecta 1 0 2 L 
Creiis longipennis 1 0 2 L 
Spondyliaspis plicatuloides 1 0 2 L 
Cardiaspina albitexturata 1 0 2 L 
Cardiaspina vittaformis 1 0 2 L 
Glycaspis baileyi 1 0 2 L 
Glycaspis aggregata 1 0 2 L 
Glycaspis planitecta 1 0 2 L 
Ctenarytaina eucalypti 0 0 1 F (W) 
Colophorina cassiae 0 0 1 F (W) [between pairs of unopene leaflets] 
Euphalerus sp. A ? ? ? unknown 
Euphalerus nidifex 1 0 2 L 
Russe lliana adesmiae 0 0 0 F 
Russelliana fabianae 0 0 0 F 
Heteropsylla texana 0 0 0 F 
Acizzia uncatoides 0 0 0 F 
Trigonon longi ? ? ? unknown 
Psylla alni 0 0 0 F 
. See text for the details . 
.. F = free-living; W = wax; L= lerp; P= partial gall (not enclosed); G = enclosed gall; ? = biology unknown; W within 
parentheses means associate habits with F; notes are enclosed in bracket. 
, 
...... 
°' N 
~, leaf galls leaf galls 
__,..--...._ 
Blister gall 
-
-~ Star gall 
0 glabrous Nipple gall , 
"'- r 
' 
I 
r \T~ig !!:'" \p-1 ) \ u hairy Nipple gall 
~ Disc gall 
Fig.1-1. Diagrammatic profiles of gall types of hackberry psyllids, Pachypsylla, on 
host plant Ce/tis. There are four major groups according to galling position, i.e., the 
leaf-blade gall makers, the petiole gall makers, the bud gall makers, and the twig gall 
makers. Drawings in the box show five major types of leaf galls . 
....... 
°' \..,.) 
Figure 1-2. Distribution of hosts of the hackberry psyllids, Pachypsylla (after Little, 1971, 1976, 1977) 
,......,. 
°' ~
NO.TH DAKOTA 
MONTANA I SOUTH OAJ:OTA 
HEUASl:A 
WYOMING 
lA~SAS 21 
All ZONA 
COlOUoo 
TUAS 
1 
HfW M.f.XICO 2 
Figure 1-3. Map of United States showing sampling localities for Pachypsylla adults. 
Numbers of populations sampled from each state are indicated. 
-O'\ V\ 
NOUH o.o :o rA 
1 
MO N TAN A 
SOUTH OA k O TA 
Mf llA Sl A 
WYOM ING 
lAHSAS 11 
COlOIAD_O 
HIW M.f.XICO 
.H"' 
Figure 1-4. Map of United States showing sampling localities for Pachypsylla nymphs. 
Numbers of populations sampled from each state are indicated. 
,..... 
0\ 
0\ 
(d) 
~ 
Figure 1-5. Forewing pattern variations of the major galling groups of 
Pachypsylla adults: (a) the petiole gall maker, P. venusta; (b) the leaf-
blade hairy nipple gall maker, P. celtidismamma; (c) the glabrous bud gall 
maker, P. celtidisgemma; (d) the hairy bud gall maker, P. pallida; (e) the 
twig gall maker, P. celtidisinteneris. (after Porter) 
f--' 
0\ 
-.J 
(a) 
~-«-<-<-<M .. . . . ..... . 
(b) 
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC 
Figure 1-6. Life history of two types of leaf gall makers at National Agricultural 
Library (NAL) population in Beltsville, Maryland: (a) glabrous nipple gall 
maker; (b) star gall maker. 
(a) Leaf blister gall maker 
• 
(c) Leaf hairy nipple gall maker ( d) Side cell individual of leaf 
hairy nipple gall 
,__.,. 
°' 
'° 
~ 
(a) 
Wagner tree produced by rooting at midpoint of longest path 
Coefficient used: Rogers (1972) genetic distance 
w********************************************************** 
Di s tance from root 
. 00 .01 .02 .03 .04 .D5 .06 .06 .07 . 08 .09 
+----+----+----+----+--- -+- ---+----+----+----+----+--- -+--- -+----+----+----+- - - -+ ----+--- -+- ---+----+ 
Hairy Nipple Gall rn+c (CMt) 
~--- - -------- Glabrous Nipple Gall rn+c (NAL+BB) 
.-- ---------------- - - ---------- - --- Blister Gall m (CMt) 
Star Gall m (NAL+BB) 
.-------------- ----------------- Hairy Nipple Gall side (CMt) 
'--------------- Glabrous Nipple Gall side (NAL+BB) 
+- ---+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+- - --+----+----+----+ 
.00 .01 .02 .03 .04 .05 .06 .06 .07 . 08 .09 
(b) 
Cluster analysis using unweighted pair group method (UPGMA) 
Coefficient used: Rogers (1972) genetic distance 
*****************'************************************ 
Distance 
.20 .18 .16 .14 . 12 .10 .08 .06 .04 .02 .00 
+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+--- -+----+----+- -- -+----+ 
~----------- --- -- Hairy Nipple Gall m+c(CMt) 
Glabrous Nipple GAll rn+c (NAL+BB) 
'-------------------- Star Gall m (NAL +BB) 
'--- -------------------- Glabrous Nipple Gall side (NAL+BB) 
'---- -------------------- ---- Blister Gall m (CMt) 
'-- -------------------------- Hairy Nipple Gall side (CMt) 
+----+----+----+----+----+----+----+----+----+----+----+----+----+----+----+- -- -+----+----+----+----+ 
.20 .18 .16 .14 .12 .10 .08 .06 .04 .02 .00 
Figure 2-1. Distance Wagner and UPGMA trees, based on Rogers' distance (15 loci), of eight populations of 
Pachypsylla leaf gall nymphs from different gall types and cell positions (m= monocell gall; c= center cell of multiple 
cell gall; side= side cell of multiple cell gall) . Populations are given in parentheses; acronyms follow Table 1-3. 
411 _____ _ 
,,/ ,I 3/3 (a1 
3 
2 
/ , 
, 
, 
, 
, 
, 
, 
, 
, 
, 
, 
,r' /l·t-----lI 0/4 
or=='==========='=== 
A--------I (b) 
4 
3 
2 
, 
I / 
y 
, 
, 
/ 
,,, 
,,)·/ 13/15 
3/7 
or================ (c) I 
I 
5 
,//F--------r 5/5 I 
4 I 
--I 5/9 
3 
2 
Observation time 
-- dest royed 
---- · und is turbed 
Figure 2-2. Change of gall size through time betw 
individual destroyed) and control (center cell indivr~n ratll!ent (center cell 
was conducted at three different times during gall i·n·tu_a _undisturbed). Destruction 1 iat10n · (a) b th 
cell nymphs exposed, (b) center cell nymph enclosed b t .d O center and side 
(c) both center and side cell nymphs enclosed in galls ~i1 e ce!l nymphs exposed, 
when destruction implemented; Time 2: two weeks after ti servtio_n time I : time 
after time 1; Time 4: fou r months after time I. Number off~- ·.Jime 3: two months 
total observed indi viduals until emergence is given at time 4~ ivi ual survived out of 
170 
w 
~ 
<( 
I-
z 
w (..) 
,-....,. cc 
--....) w 
,-....,. a.. 
100 
80 
60 
40 
20 
0 
4/15 
(55) 
4/20 
(65) 
- Center cel l female ~ Side cell female 
4/25 
(70) 
4/30 
(70) 
DATE 
5/5 
(62) 
5/10 
(56) 
5/15 
(50) 
Figure 2-3. Change of proportion of center and side cell females of hackberry 
leaf gall makers at Beltsville, Md, in the spring of 1993. Numbers in 
parentheses below each date indicate total number of individuals collected. 
(a) 
!!ft• 
3
-1. Female genitalia of: side cell individual of Pachypsylla hairy nipple 
(d) !a), s!3r gall maker(b), Tetragonocefhalajlava (c) and Ce//isaspis beijingana 
pr ' how mg rounded dorsal valve (DV) ID the latter two taxa, and thumb-like 
ocess (arrow) in the former two species. 
172 
-
--.J 
v-) 
Distance from root 
. 00 .06 .11 . 17 .22 .28 .33 .39 .44 . 50 .55 
+- - - -+-- - -+-- - -+-. - -+- -- -+- - - -+- - - -+- - - -+- - --+- - --+- - --+- - --+- - - -+- - - -+- -- -+-- - -+-- --+- - - -+- - . -+----+ 
dlstance \.Jagner tree 
Rogers' di stance 
tree length = 2.849 
Hairy Nipple Galler mono+center (CMt) 
Disc Ga l ler mono (GFV) 
Hairy Nipple Galler side {CMt) 
Glabrous Nipple Galler side {NAL+BB) 
Star Galler mono (MD) 
Blister Galler mono {CMtl 
Bli s ter Galler mono {GFV310) 
Blister Galler mono (OH+ND+ID) 
Blister Galler mono Cl (TX) 
Blister Galler mono C.r. (TX) 
Glabrous Nipple Galler mono+center {NAL+BB) 
Glabrous Nipple Galler mono {GFV) 
Petiole Galler(BFL) 
Petiole Galler {BB) 
Petiole Galler {AZ) 
Glabrous Bud Galler {LA) 
Glabrous Bud Galler {NAL+BB) 
Hairy Bud Galler {OK) 
G l abrous Bud Gall er {VA) 
~ - -------------- -------- Twig Galler {OK) 
Te tr agonocepha la fl ava 
~----------------------- -------- Cletisaspis bei j ingana 
+- - - -+-- - -+----+-- --+- - - -+- - - -+- - --+- ---+- -- -+- - - -+-- --+- - - -+- - - -+- - - -+- - -+- - - -+- ---+ 
.00 . 06 .11 . 17 .22 .28 .33 .39 .44 .50 .55 
Figure 3-2. Distance Wagner tree of Paclzypsylla with outgroups Tetragonocephala flava and Celtisaspis beijingana 
based on Rogers' genetic distance using allozyme data (15 loci). 
,...... 
-..J 
~ 
1.00 .90 .80 . 70 .60 
Distance 
.50 .40 .30 . 20 .10 .00 
+- ---+--- -+- - - -+- - - -+- - --+-- - -+- - - -+-- - -+- - - -+- - - ·+-. - -+- -
(al 
UPGMA tree 
Rogers' distance 
CPCC=O. 986 
******************* 
****• Hairy Nipple Galler mono+center (CMt) 
***** Disc Galler mono (GFV) 
* ******** G l abrousN 1 ppl eGa l lermono+center ( NAL +BB) 
Glabrous Nipple Galler mono (GFV) 
Star Galler mono (MD) 
•••••••••** Ha 1 ry Nipple Gal ter side ( CMt) 
*********** Glabrous Nipple Galler side (NAL+BB) 
** *** * * * * * * ** * * * * * * * * * * ********** * * * * * *** * * * * *. * 
Blister Galler mono (CMt) 
Blister Galler mono Cl (TX) 
Bl ister Galler mono Cr (TX) 
*********** Blister Galler mono (GFV310) 
*********** Blister Galler mono (OH+ND+ID) 
* *** * ** * ** * * ••• *** •• ************* *** ** *** * * ** * ** * **** **** ****** Tetragonocephala flava (AZ) 
Glabrous Bud Galler (LA) 
Glabrous Bud Galler (MAL+BB) 
****************************************************** Hairy Bud Ga l ler (OK) 
********** ******** Glabrous Bud Galler (VA) 
****************************************************** Twig Galler (OK) 
Petiole Galler (BFL) 
•••••• **. *. *. * * *. ** **** * * *** •• ** ** •• ********* * *. * •• * ••••••••• Petiole Galler (BB) 
Petiole Galler (AZ) 
** •• *** **** ••••••• * * *** * * * * * ** * *** *** ** ••••••• ***** •• * * * •• ***. * ** *. * •• *. **** •••••• **** Celtisaspis beijingana (CHI) 
+- -- -+- - - -+- - - -+--- -+-- - -+- - --+- - - -+-- - -+- - - -+- - - - +----+- - - - -+- -- -+- - --+-- --+- - - -+-- - -+ 
1. 00 .90 .80 .70 . 60 . 50 .40 .30 .20 . 10 .00 
Figure 3-3 (a) UPGMA tree of Pachypsylla with outgroups Tetragonocephala flava and Celtisaspis beijingana based on 
Rogers' genetic distance using allozyme data (15 loci). 
l 
........ 
--.) 
VI 
S imil arity 
.00 .10 .20 .30 .40 . 50 .60 .70 . 80 .90 1.00 
+- - --+----+- - - -+ - - - -+ - -- -+-- - -+ - - --+ - - - -+----+ - -- -+- - - -+- - - -+ - - - -+--- -+-- --+----+- - - -+ - -
-+ 
(b) 
UPGMA tree 
Nei 1 s distance 
CPCC= 0.988 
* Hairy Nipple Galler mono+center (CHt) 
Di SC Galler mono (GFV) 
***G l abrousN i ppleGa l lenmono+ce nter(NAL +BB) 
*** 
**** G\abrous Nipple Gal \e r mono (GFV) 
**** 
* •••• Star Galler mono (MD) 
* * **** Glabrous Nipple Galle r side (NAL+BB) 
• * 
•******Hairy Nipple Galler side (CMt) 
***************** .............................. ********** 
** Blister Galler mono (CMt) 
*** 
* ** Blister Galler Cl (TX) 
• Blister Galler Cr (TX) 
**** Blister Galler mono (GFV310) 
*** Blister Galler mono (OH+ND+!D) 
* **************'*************************"'*************'********** T etragonocephal a f lava {AZ) 
..... 
** Glabrous Bud Galler Cl (LA) 
** 
* ***- Glabrous Bud Galler Ct (NAL+BB) 
****************************** .. ********************** Hairy Bud Galler (OK) 
...................... • * 
**** Glabrous Bud Galler Co (VA) 
••••••••••••••••••••11r•••••••••••••••••••••••••••••••• Twig Galler (OK) 
**** Petiole Galler (BFL) 
......... 
**********************•*********************************** **** Petiole Galler (BB) 
************ Pet i ote Galler CAZ) 
****************************************************************************************** Celt i saspi s bei j i ngana (CH I) 
+----+----+----+----+-----+----+----+----+----+----+----+----+----+----+----+----+---
.00 . 10 .20 .30 . 40 .50 .60 .70 .80 .90 1.00 
Figure 3-3 (b) UPGMA tree of Pachyp~ylla with outgroups Tetragonocephala flava and Celtisaspis beijingana based on 
Nei's genetic distance using allozyme data (15 loci). 
Leafgall .-----
making 
complex 
Celtisas is 
Tetra onoce /za[a 
P e all (center) MD Hairy Nip I G 
Hairy Nipple Gall (side) MD 
Glabrous Ni 1 pp e Gall (center) MD 
ipp e Gall (center) VA Glabrous N' 1 
Glabrous Nipple Gall (side) MD 
Star Gall (center) MD 
Blister Gall (center) MD 
Blister Gall (center) VA 
Blister Gall (center) OH+ND+ID 
Blister Gall (center) 1X Cl 
Blister Gall (center) 1X Cr 
Disc Gall (center) VA 
Petiole Gall 1X 
Petiole Gall MD 
Petiole Gall AZ 
Glabrous Bud Gall LA 
Glabrous Bud Gall MD 
Glabrous Bud GAIi VA 
Hairy Bud Gall OK 
Twig Gall OK 
Celtisaspis beijingana 
Tetragonoceplzala jlava 
F· tgure 3 
altoz -4. Strict 
stat Yrne dat consensus of 3400 most parsimonious trees for nymphal 
C es. Cell Pa ~hen loci treated as characters alleles as discrete character 
· reli os1f ' 
cufata. ion within a gall in parentheses. Cl: Ce/ti< faevigata; Cr: 
176 
---
Leaf gall 
making 
complex 
Hairy Nipple Gall (center) MD 
Disc Gall (center) VA 
Glabrous Nipple Gall (center) MD 
Hairy Nipple Gall (side) MD 
Glabrous Nipple Gall (side) MD 
'---- Star Gall (center) MD 1.o.----- Blister Gall (center) VA 
a.------ Blister Gall (center) MD i--------- Blister Gall (center) 1X Cr i---------- Blister Gall (center) 1X Cl 
Glabrous Nipple Gall (center) VA 
Blister Gall (center) OH+ND+ID 
Pachypsy/la Petiole Gall 1X 
petiole Gall MD 
Pachypsyllini 
'---- petiole Gall AZ 
Glabrous Bud Gall LA 
Glabrous Bud Gall MD 
Glabrous Bud GAll VA 
flairY Bud Gall OJ{ 1---- TWig Gall OJ{ 
L----_:::.C::el:ti::,·s::;as ~p:,:i::.,s ________ Celt is asp is beijinga na 
L----,:_T;e:.:_tr,:a!.g:on :o~c:;ep~h :,::a;;I;a_______ retrago nocep ha/a Jlava 
Figure . onious trees for .uozyine 
data _3-5. Strict consensus of four most p•rs,in resentf•bsent cb•racter. 
Cell wit~ _each allele treated as an indepen~;.1/ /tis [aevigata; Cr: C. 
relic position within a gall in parentheses• • e 
ulata. 
177 
(a) 
(c) 
Blister Gall MD 
Blister Gall TX 
Star Gall 
Glabrou N" s ipple Gall (center) 
Glabrous Nipple Gall (side) 
Hairy Nipple Gall (side) 
Hairy N" ipple Gall (center) 
Disc Gall 
Petiole Gall 
Glabrous Bud Gall 
Hairy Bud Gall 
Twig Gall 
Tetrago nocephala Jlava 
Celtisaspi b ... s e1J1nga11a 
Blister Gall MD 
Blister Gall TX 
Star Gall 
Glabro N" us ipple Gall (center) 
Glabrous Nipple Gall (side) 
Hairy N" ipple Gall (side) 
ipple Gall (center) Hairy N" 
Disc Gall 
Petiole Gall 
Glabrous Bud Gall 
Hairy Bud Gall 
Twig Gall 
Tetrago nocephala Jlava 
Celt is · aspis beiji11gana 
(b) 
(d) 
Blister Gall MD 
Blister Gall 'IX 
Star Gall 
Hairy Nipple Gall (side) 
Glabrous Nipple Gall (side) 
Glabrous Nipple Gall (center) 
Hairy Nipple Gall (center) 
._..-- DiscGall 
--- Petiole Gall 
Glabrous Bud Gall 
Hairy Bud Gall 
Twig Gall 
Tetrago11ocepl,alaf1ava 
Celtisaspis beijingana 
--- Blister Gall MD 
._.-- Blister Gall 'IX a.--- Star Gail Glabrous Nipple Gall (center) 
GJabrous Nipple Gall (side) 
Hairy Nipple Gall (side) 
Hairy Nipple Gall (center) 
Disc Gall 
--- petiole Gall 
Glabrous Bud Gall 
Hairy Bud Gall 
Twig Gall 
Tetragonocepliala jlava 
Celtisaspis beijingana 
l?igu 
for P. te 3-6. Th Pos117t Yf sy/far':;' equally parsimonious trees ( a-c) ,.;;d their strict cons:~J t) 
n •n par ased on morphological data with all characters unorder · 
en theses. 
178 
(a) 
(c) 
Blister Gall MD (b) 
,-- Blister Gall 1X 
Star Gall 
ipple Gall (center) Glabrous N" 
us ipple Gall (side) Glabro N" 
ipple Gall (side) Hairy N" 
ipple Gall (center) Hairy N" 
Disc Gall 
Petiole Gall 
Glabrous Bud Gall 
Hairy Bud Gall 
Twig Gall 
Tetrag onocephala jlava 
e1;inga11a Celtisaspis b . .. 
. Blister Gall MD 
.._ Blister Gall 1X 
Star Gall 
Glabrou N" s ipple Gall (center) 
Glabrous Nipple Gall (side) 
Hai ry Nipple Gall (side) 
ipple Gall (center) Hairy N" 
Disc Gall 
Petiole Gall 
Glabrous Bud Gall 
Hairy Bud Gall 
Twig Gall 
Tetrag onocephala jlava 
s e1;111ga1ia Celtisaspi b ... 
(d) 
Blister Gall MD 
Blister Gall TX 
--- ~arGall 
GJabrous Nipple Gall (center) 
Glabrous Nipple Gall (side) 
I{airy Nipple Gall (side) 
I{airy Nipple Gall (center) 
--- Petiole Gall 
Glabrous Bud Gall 
Hairy Bud Gall 
Twig Gall 
Tetragonocepha la fla va 
Celtisaspis beijingana 
--- Blister Gall MD 1---- Blister Gall TX 
~arGall 
Glabrous Nipple Gall (center) 
Glabrous Nipple Gall (side) 
I{airY Nipple Gall (side) 
I{airY Nipple Gall (center) 
L---- petiole Gall 
GJabrous Bud Gall 
I{airY Bud Gall 
Twig Gall 
Tetragonocephala jlava 
Ceftisaspis beijingana 
l?igu 
for re 3-7 
~ Pach • Three d h · tricl consensus ( d) 
"l'etith YPsy/fa m equally parsimonious trees (a-c) an trde1r J Cell position in 
eses. orphological data, with all characters 
O 
er • · 
179 
~ 
,_. 
00 
0 
Figure 3-8. Scanning electron micrographs of adult heads: (a) Pachypsylla leaf star galler; 
(b) Pachypsylla petiole galler; (c) Tetragonocephalaflava; (d) Celtisaspis beijingana, 
showing the surface structure of vertex (V) and genae (G). 
(b) 
(a) 
Hairy Nipple Gall (center cell) 
Hairy Nipple Gall (side cell) 
Glabrous Nipple Gall (side cell) 
t----- Glabrous Nipple Gall (center cell) 
Disc Gall (center cell) 
Star Gall (center cell) 
r------- Blister Gall (center cell) MD 
Blister Gall (center cell) TX 
r-------- Petiole Gall 
Glabrous Bud Gall 
Hairy Bud Gall 
Twig Gall 
Celtisaspis beijingana 
Tetragonocephala jlava 
,------ Hairy Nipple Gall (center cell) 
Hairy Nipple Gall (side cell) 
Glabrous Nipple Gall (side cell) 
---- Disc Gall (center cell) 
----- Glabrous Nipple Gall (center cell) 
------- Star Gall (center cell) 
r--------- Blister Gall (center cell) MD 
--------- Blister Gall (center cell) TX 
------ Petiole Gall 
Glabrous Bud Gall 
Hairy Bud Gall 
---- Twig Gall 
------------ Celtisaspis beijingana 
------------ Tetrogonocephalajlava 
};'· k igure 3-9 
cttyotypi · ;ons~nsus trees for combined morphological, el~ctrophoretic, 
C torn JO t c, nd hfe history data: ( a) each locus treated as single character 
elJ Posjftee.s); (b) each allele treated as single character (from 5 trees). 
ion of leaf galls in parentheses. 
181 
..... 
00 
N 
Figure 3-10. Fifth instar nymph of leaf blister gall maker: (a) dorsal view; (b) ventral 
view; (c) enlargement of caudal spurs (arrow) of (a); (d) enlargement of caudal spurs 
(arrow) of (b ). 
>---' 
00 
v.) 
(d) 
Figure 3-11. Hind coxa of hackberry leaf star gall maker showing: (a) distal ventral crescentic denticulate 
patch (arrow head) and a fan-shaped area (arrow) near the patch, ventral view; (b) lateral view; (c) 
enlargement of denticulate patch of (b); (d) enlargement of hole (double arrow head) between fan-shaped 
area and trochanter (T) of (a). 
.. 
:, 
Q. 
~ 
Q. 
5' 
"' C 
:, 
other psyllids 
(Calopsylidae, Horroto . . 
rr,dae, Phacopteron1dae. Carsidaridae and Triozidae) 
Aphalarinae (Aphalaridae) 
Rhinocolinae (Aphalaridae) 
Diaphorininat 
Paumctphalinat 
.-------- Diclidophltbiini 
.-----Euphyllurini 
Eucalypro/y""1 
-==-----------------~== CttnarYtaina - 1,phaloroila ,4.r,puna (dlussd/iana)- Arepuriln•• 
Colopt,arina J Euphol ip. A E14>18ltrl ,.... 
Euphal spp. J. R. V. 
Pochypsy//a 
r,,ragonactphala 
CellisaSPis 
f/1Lllopsylla 
Euphaltr nidi[u 
Euphal ga//ico/us 
carrJiaspina 
J Pachyp,yllln•• 
]-"·'"' 
Trigo nan 
Mirrapsy//a 
A,cizzi• 
Neopsy/la P/orycof)'Pha 
Fnysuila 
Anomontura 
Epipsy//a 
Euphal sp. C&D 
Euphal sp.B 
t,""1phicola & Eupha/,rus ranr/1/US 
Heterapsylla 
Euceropsy//a 
Jsogonoceraia 
/nJTltsiO 
CiriacnttrJITI 
A,ryroina 
Floria 
A,ryroini//a 
Psylla (Ba,opt/""1) 
Span/ontUrD 
psylla (Asphagidtlla) 
psy//a (.str.) 
pry/lo (Heparopry//a) 
pry/la (OYIIOPSJ11af7honviopry11a) 
Cennothia 
Euglypron,ura 
fur.vtivora 
pry/la (Cacop.ry//a) 
};" 
._!gure 4 Q.ighI · · l. Parr 1 
gh led tax~•! P ~YI ogen y of Psyll oi dea after While and Hodkinson ( 1985), 
md1cate those sampled for current study, 
184 
a 
b 
C 
Fig. 4-2. Shapes of occipital foramen. 
185 
(a) 
(c) 
Pachypsyl/a ce/tidismamma 
Pachypsy/la cellidisvesicula 
Pachypsy/la venusra 
Pachypsy/la cellidisgemma 
Te rragonocephala j/ava 
Ce!ri.saspis beijingafla 
Creiir tecra 
Cardiaspina a/birexrurara 
Cardiasp1na vinaformis 
Creiis lrmgipennis 
G!ycaspis baileyi 
G!ycaspis aggregara 
Glycaspis planirecra 
Spondyliaspir p/icaruloid 
Crenaryraina eucalypri 
Phellopsy/la spp. 
Phyllolyma rufa 
Phyllo!yma sp. 
Colophorina sp. n 
Euphalerus sp. A 
Euphale rus nidifix 
Hereropsy//a rexana 
Trigonon lnngicornis 
A c iwa uncaroides 
Psy/la a/ni 
Russel/iana fabiahae 
Russel/iana adesmiae 
Pachypsylla ce/tidismamma 
Pachyp.rylla ce/1id1:rvesicula 
Pachypsyl/a venusra 
Pachypsylla celridirgemma 
Te lragonocepha/a jlava 
Ce lrisaspir be ijingana 
Creiir re era 
Creiir longipeflflir 
Cardiasp1na a/hirexrurata 
Cardiaspina vinafonnis 
G!ycaspir baileyi 
G!ycaspir aggregata 
G!ycaspis planirecra 
Spondyliaspis plicaru/oid 
Crenarytmna eucalypti 
Phel!opsylla spp. 
Phyl!olyma rufa 
Phyllolyma sp. 
Colophorina sp. 11 
Euphalerus sp. A 
Euphalerus nidifu: 
Hereropsyl/a rexa11a 
Trigo11011 I011gicornis 
Aciu.ia 1111caroides 
Psylla a/11i 
Russe llia11a fabiahae 
Russe//ia11a adesmiae 
(b) 
(d) 
Pachypsy//a celridirmamma 
Pachypsy//a ce/Jidirvesicula 
Pachypsy//a ve11us1a 
Pachypsy//a ce/Jidirgemma 
Te tragonoce phala jlava 
Ce lrisaspir be ijinga11a 
Creiir lecra 
Creiir longipennir 
Cardiaspina alhirexturata 
Cardiaspina vinafonnir 
G/ycaspir baileyi 
Glycaspir aggregara 
Glycaspir planilecra 
Spondyliaspir p/icarulo_id 
Cre11aryraina euca/yplt 
Phe//opsylla spp. 
Phy//olyma rufa 
Phyllo/yma sp. 
Co/ophorina sp. n 
Euphalerus sp. A 
Eupha/erus mdifu: 
Hereropsylla rexana 
Trigonon /011gicomir 
Aciuia u11catmdes 
Psyl/a alni . 
Russe/lia11G fab,alzae 
Russe Iliana adesmiae 
Pachypsy//a cellidi.rma,~ma 
Pachypsylla ce/ridisves,cula 
Pachypsylla venusta 
Pachypsy//a celridisgemma 
Terrago110cep.~a/ajlava 
Celrisaspir bel}mgana 
Creiis recto 
Creiis lo11gipen11ir 
Cardiaspina alhitexrurata 
Cardiaspi11a vinafonnis 
Glycaspir ba11eyi 
Glycaspir aggregata 
G/yca.rpir p~amtuta . 
Spondyliaspis p/,catulmd 
Crenarytai11a euca/yplt 
Phe//opsyl/a spp. 
Phy//olyma rufa 
Plzy//o/yma sp. 
Colophorina sp. 11 
Euphalerus sp. A 
£up hale rus 11idifix 
Heteropsy//a texa11a 
Trigonon fongicom1s 
Aciuia uncato1des 
Psyl/a alni . 
Russe Iliana fabwlwe 
Russellia11a adesmwe 
}?· 
t Jgllte 4-3 d their strict consensus ( d) 
S •t SpOlJd . th tee equally parsimonious trees ( a-c) an. th all characters unonlered. 
allle tr: y Jaspidini based on morphological data wi 
ees found after successive weighting. 
186 
Pachypsylla celtidismamma 
Pachypsylla celtidisvesicula 
Pachypsylla venusta 
Pachypsylla celtidisg , ma 
Tetragonocephala ftava 
Celtisaspis beijingana 
Creiis tecta 
Creiis longipennis 
Spondyliaspis plicatuloid 
Cardiaspina albitexturata 
Cardiaspina vittaformis 
Glycaspis baileyi 
Glycaspis aggregata 
Glycaspis planitecta 
Ctenarytaina eucalypti 
Phellopsylla spp. 
Phyllolyma rufa 
Phyllolyma sp. 
Colophorina sp. n 
Euphalerus sp. A 
Euphalerus riidiftx 
Heteropsylla texana 
Trigonon [ongicornis 
Acizzia uncatoides 
------------- Psylla alni 
Russelliana Jabiahae 
Russelliaria adesmiae 
F· S •gure 4-4 . Pondyl' · ~tnct consensus of 164 most parsimonious trees for 
ch 1asp1d' · · h aracters # mae based on morpbOlogical data wit some 
( 3-7) ordered and others unordered, 
187 
Pachypsylla celtidismamma 
Pacliypsylla celtidisvesicula 
Pachypsylla venusta 
Pachypsylla celtidisgemma 
Tetragonocephala Jlava 
Celtisaspis beijingana 
Creiis tecta 
Creiis longipennis 
Cardiaspina albitexturata 
Cardiaspina vittaJormis 
Spondyliaspis plicatuloid 
Glycaspis baileyi 
Glycaspis aggregata 
Glycaspis planitecta 
Cteriarytaina eucalypti 
Phellopsylla spp. 
Phyllolyma rufa 
Phyllolyma sp. 
Colophoriria sp. n 
Euphalerus sp. A 
Euphalerus riidifix 
Heteropsylla texaria 
Trigonori /o,igicorriis 
Acizzia uricatoides 
Psylla alni 
Russelliana Jabiahae 
Russelliaria adesmiae 
F· igure 4-5 . 
8
Pondy(' • Stnct consensus of 10 most pars;n1onious tree' for 
(#3-7) iaspidinae after successive weighting with some characters 
and others unordered, ordered 
188 
REFERENCES 
Abe, Y. 1992
· Ethological isolation between inquiline gall wasps, the Synergus 
Japonicus 
complex (Hymenoptera: Cynipidae ), which sometimes coexist in 
host g 11 l\ki a s. Appl. Entomol. Zool. 27(4): 527-531. 
I11oto, s. 1981. Gall formation by Erisoma fundatrices and gall parasitism in 
Erioso,n . 
ayangz (Homoptera, Pemphigidae). Kontyu, To.kyo, 49(3): 426-436. 
---------- ( 
--. l988 Co · · · h' E · h'd 
· mpet1t10n and niche relatwns 1ps among nosoma ap 1 s 
occurrin 
I\ g on the Japanese elm. Oecologia (Berlin) 75: 44-53. 
lbertson F 
' · W. annd J. E. Weaver. 1945. Injury and death or recovery of trees 
in Prairi 1 · An e c 1mate. Ecological Monographs, 15: 394-443. 
derson, S . . 
· and C. G. Tauer. 1993. Hackberry seed sources for planting m the 
southern G · 44(2) 
reat Plains. Tree Planters' Notes, USDA Forest Serv1ce, : 
78-82. 
Ash Ill 
ead, W. H. 1881 . On the Aphididae of Florida with descriptions of new 
Sp . 
ecies (Family Psyllidae). Canadian Entomologist, 13: 220-225· 
Askew, R . . , 
· R. 1961. On the biology of the inhabitants of oak galls of Cyrnpidae 
(Bynienoptera) in Britain. Trans. Soc. Br. Entomol. 14: 237-268· 
------------. 1980 T'h . . . · 1 f miners and plant galls. 
· e diversity of insect commumt1es m ea - c 
J. Anim. Ecol. 49: 817-829. 
------------. 1984 The b · 1 f 11 223 
?73 Jn Ananthakrislman, T. N. 
· 10 ogy o ga wasps. pp. -- · 
[eel.] Biology of gall insects. New Delhi: Oxford & IBH Pulbishing Co. 
189 
A.skew R 
' . · R. and J M R . . .. 
· · use. 1974. The bwlogy of some Cec1domy11dae (Diptera) 
galling tl 1 Je eaves of birch (Betzda) with special reference to their 
chalcidoid (H 
ymenoptera) parasites. Trans. R. Entomol. Soc. London. 126: 
129-167. 
Beardsley J W 
' · · 1987. Overview of Heteropsylla cubana, Leucaena and integrated 
Pest management: Psyllidae or jumping plant lice: Note on biology and 
control L 
· eucaena Research Reports, Proceedings of a workshop on the 
Biologi 1 , . . . . 
ca and Genetic Control strategies for the Leucaena psyll1d. Special 
issue, 7(2): 1-5. 
Becker M· 
- igdisova, E. E. 1949. Mezozojskie Homoptera Srednej Azii. Tr. PIN, 
Moskva, 22: 1-68. 
--- ........ ___ _ 
----. 
1973
· On the systematics of the Psyllomorpha and the position of the 
group · h · D kl d Wit m the order Homoptera. In Narchuk, E. P. [ed.] o a Y na 
dvadzat chetvertom escheghodnom chtenii pamyati N.A. Kholodovskogo. 
1971 L · · · B .. h 
· enmgrad, 1973: 90-117. [In Russian, Engllsh translatwn - nt1s 
Library B 
, oston Spa.] 
------------· 1985 I . T PIN Moskva 206: 1-92. 
13 . · skopaemye nasekomye ps11lomorfy. r. , · , 
e1s/er, J. i\,,. 'd' 
1v1 1989. Morphology of sensilla of adult Paclzypsylla celtt LS!namma 
(Hornoptera: Psyllidae) and of the gall formed on leaves of Ce/tis 
occidentalis (Urticales: UJmaceae). pp.964-965. In Bailey, G. W. [ed.] 
Proceedings of the 47th Annual Meeting of the Electron Microscopy 
Society f . . 
o Amenca. San Franc1sco Press. 
190 
B· e1sler, J. M. and G: T. Baker. 1992. Pachypsylla celtidismamma (Fletcher) 
(Homoptera.· p 11·ct sy 1 ae): Morphology and history of its gail and 
ultrastruct . B ure of its adult and nymphai sensiila. Miss. Ent. Mus. No.3. 27pp. 
olton B , . 1986 A tax . d . . . f h .. · anomic an b10log1cal review o t e tetramorune ant genus 
Rhoptr omynnex (Hymenoptera: Formicidae). Syst. Entomoi. !1(1): 1-17. 
Boselli, F. B . . . . · l929. Studu sugli Psyilid (Homoptera: Psyi11dae o Cherm1dae). !II. 
Appunti su alcune nimfe di Pachypsyilini. Boil. Lab. Zool., Portici 22: 204· 
218. 
Bourke F F ' · · G. and N. R. Franks. 1991. Alternative adaptations, sympatric 
speciatio l . · · · · · B' J J L' S c n am the evoiut10n of paras1trc, mqu11tne ants. 10 · · mn. 
0 
· 
43 (3): 157-178. 
Brookfield J . . . . , ' · F. 1972. The inhabitants (Hymenoptera: ey01p1dae, Chalcido
1
dea) 
of the cynipidous gails of Quercus borea/is in Nova Scotia. Can. Entomol. 
104: 1123-1133 B . 
rown, R. W. 
1956 c ·t· f · t'f'c worcls 882pp Smithsonian Inst. 
· omposr 10n o sc1en 1 1 · · 
Press W 
13 
', · ashington, D. C. 
rown R . . 
. and I. D. Hodkinson. 1988. Taxonomy and ecology of the Jumping plant· 
lice f p 0 anama. 
Bure],] 
'- 1ardt D 19 . p lloidea) of the temperate 
' · 87. Jumping plant lice (Bomoptera. sy 
neotropical . p 11·d e (subfamilies Aphalarinae, 
region. Part 1: sy 1 a 
Rhino . . . J L' soc. 89: 299-392. 
... colmae and Aphaiaro1dmae). zool. · 
100
· 
---
------. 1991 . ·d· classification (Hornoptera: 
· Boreioglycaspis and Spondy1Iasp1 ine 
191 
PsylJ ·d 01 
ea). Raffles Bull. Zool. 39(1): 15-52. 
Burckhardt D 
' · and Bauterer. 1989. Systematics and biology of the Rhinocolinae. 
1 Nat B· 
. . lStory. 23: 643-712. 
Bush G 
' · L. 1969 Sym t · h c · d · · · f · · 
· pa nc ost race 1ormat10n an spec1at10n rn rug1vorous flies 
C of the genus Rlzagoletis (Diptera, Tephritidae). Evol. 23:237-251. 
aldweu J 
' · S. 1938. The jumping plant-lice of Ohio (Homoptera: Chermidae). 
Oh· 10 Bio. Bull. 34: 228-281. 
Can1pbe]J K . . 
' · G. 1964. Sound production by psyJ11dae (Hem1ptera). J. Ent. Soc. 
Aust. (N.S .W.) 1·"' -4 
' ,.) . 
Capener, A. L. 1973. Southern Africa Psyllidae (Homoptera). 3. A new genus and 
new sp · f · 36· 37 ecies of South African Psyllidae. J. ent. Soc. southern A nca · -
61. 
Carver, M I . . . , . 
· 
987. D1stmctive motory behavior in some adult psyllids (Homopterd. 
Psy]J · 
' 
01 dea). 1. Aust. ent. Soc. 26: 369-372. 
Corren 
' D. S. and M. c. Johnston. 1970_ Ce/tis L. pp. 492-494. Manual of vascular 
Plants of Texas. Contributions from Texas Research Fundation. A series of 
botanical d' 
stu 1es. Vol.6. 
Cr" 
ctWfo d C ·1· ) 
r ' D. L. 1911. American Psyllidae IV (A partial revision of sub1am1 ies . 
Porn 00a 1. Entomol. 3(2): 480-503. 
--
---
-------. 1914 . p . llidae of the new world. 
· A monograph of jumping plant-lice or sy 
l3nlI. D. S. Mus. 85: 1-183 . 
.............. 
---- 7 "'_,, 361 
---. 
1920. The psyllidae of Borneo. Philippine J. Sci. 1 : .)).)- · 
192 
Dobso T 
n, . 1851 0 1 
· n aap, or lerp, the cup-like coverings of Psyllidae found on the 
leaves of ce . 
rtam Eucalypts. Pap. Proc. R. Soc. van Diemen's Land, 1: 235-
241. 
Drege J 
r- auffre t F 
' · and 1. D. Shorthouse. 1992. Diversity of gall-inducing insects 
and th · 
eir galls. pp. 8-33. In Shorthouse, J. D. and 0. Rohfritsch [eds. ) 
Biology f. . 
B O msect-mduced galls. Oxford Umv. Press. 
asteaJ S 
' · and I. A. Boussy. 1987. A sensitive and efficient isoenzyrne technique 
for smau 
r-, arthropods and other invertebrates. Bull. ent. Res. 77: 407-415. 
clias T 
' · S. 1970- The genera of Ulmaceae in the southeastern United States . .T. 
Arn oJcJ A. 
'""'sboretum. 51: 18-40. 
Bvans D 
' · 
1965
· The life history and imature stages of Syne,gus paqficus 
McCracken and Egbert (Hymenoptera: Cynipidae) . Canadian 
Entom 1 . 
o og1st, 97: 185-188 . 
................. 
--. 1967 Th b. · of B osbicus mirabilfr 
· e 1sexual and agamic generatwns l,, 
(Hyme · E 99· 187-
noptera: Cynipidae), and their associate insects. Can. nt. · 
196. 
Faith D. P , . .· 
· ancI P. S. Cranston. 1991. Could a cladograrn this short have ausen by 
F':,1. ' 
" r1s ,, 
chance al . · t e Cladistics 7: l-
one?: on permutation tests for clad1st1c struc ur · ' 
28. 
J.s. 1988. HENNIG86 
' 
version 1.5. Distributed by the author, Port 
1effer · 
son Station, N. Y. 
193 
· 26. Observauons on the Chermidae (Hemiptera: Homoptera) Part Ferris, G. F 19 . 
III Ca d' · na ian Entomologist, 58: 13-20. 
Froggatt W ' · W. 1990. Australian Psyllidae. Proc. Linn- Soc. N.S.W. 25: 250-302. 
Futuyma D ' · J. 1986. Evolutionary biology, 2nd ed. Sinauer Associates, 
Sunderland, MA. 
Futuyma ' D. J. and G. Moreno. 1988. The evolution of ecological specialization. 
Ann. Rev. Ecol. Syst. 19: 207-233. 
G'l t more, M . . . . · R. 1919. Uses of plants by the Indians of the M1ssoun River reg10n. 
Thirty-third Annual Report of the Bureau of American Ethnology To the 
secreta f p . . ry o the Smithsonian Jnsitution. Washington Government nnung 
Office. 
Hanct1· Irsch A · ' · 1925. Palaeoentnmologie: Phylogenie oder Stammesgesch1chte. In 
Schroder Ch. Handbuch der Entomologie, Jena: G. Fischer Ver!., Bd 3: 
113-372. 
Bebert p . . , . D. N. and M. J. Beaton. !989. Methodology for allozyme analysis using 
cellulose , · A cti·cal l1andbook 31PP· An 
acetate electrophoresis. pra · 
educational service Helena Laboratories, Texas, USA 
Bende . . rson, G. aod R. D. Akre. !986. Biology of the myrmecophiious cncket, 
Mynne 
1 
. ll'd ) J Dans Entomol. Soc. 59(3): 
cop zzla ,nanni (Orthoptera: Gry i ae · · · 
454-467 
Besio . P-Harriso G . . n in the homopterous Psyl11dae I. 
n, . 1949. Subfamily separatIO 
Annals , d . 12 no 2· 802-810. 
an Magazine of Nat. Hist. ser. , · · 
194 
· ubfam1ly separat10n m the homopterous Psyl11dae JI. Annals and 
------------. 1951 S · . . . 
Maga· zme of Nat. Hist. ser. 12, no. 4: 1-35. 
------------. 1954· Contributions to our knowledge of the Psyllidae of Australia and 
New Zeal d . an , with special reference to Tasmania JI. Annals and Magazine 
of Nat H' · ist. ser. 12, no. 7: 519-530. 
--- 1958 
· · Subfamily separation in the homopterous Psyllidae !II (a-c). 
Annals a d M . n agazme of Nat. Hist. ser. 13, no. 1: 561-579 . 
....... ___ _ 
------. 
195
9. Subfamily separation in the homopterous Psyllidae III (d-e). 
------
1 
Annals and Magazine of Nat. Hist. ser. 13, no. 2: 157-168. 
-lodkinson, J . . . . . . · D. 1973. The b10logy of Stroplungza encae (Curtis) (Homoptera. 
PsyIIoide ) · · ! · (W lk ) a with notes on its primary parasite Tetras11c1US actIS a er 
(Hym., Eulophidae). Norsk. ent. Tidjsk, 20: 237-243. 
------------. 1980. Present day distribution patterns of the holarctic Psylloidea 
(Ho mop tera - Jnsecta) with particular reference to the origin of the nearcti c 
fauna J f . · · o B1ogeography, 7: 127- 146. 
------------. 1984. The biology and Ecology of the gaJl-forming Psylloidea 
(Homoptera). pp. 59-77. In Ananthakfishnan, T. N. [ed.] Biology of gall 
insects. New Delhi: Oxford & JBI-l Pulbisbing Co. 
------------. 1986. The psyllids (Homoptera: Psylloidea) of the Oriental 
Zoog . k 1- t J Nat flist 20: 299-357. 
eograph1cal Region: an annotated cheC is · · · ., 
-----------. 193
8 
. p ]loidea)· an annotated 
· The Nearctic psylloidea (f-Iornoptera. sy · 
check li st. J. Nat. Hist., 22: 1179-1243. 
195 
Hodkinson, I. D. and Flint, P. W. H. 1971. Some predators from the galls of 
raxmz L.) (Hem. Psyllidae). Ent. Monthly Mag. 107: 11-12. 
Psyllopsis fi · · ( 
Bodk" mson, I D . . . · · and I. M. White. 1981. The Neotrop1cal Psyl101dea (Homoptera: 
Insecta)· · an annotated check list. J. Nat. Hist., 15: 491- 523. 
Bollis, D. 1976 J . 1 . "b c· . . . (LI · umping pant bee of the tn e macrem1rn r omoptera: 
Psy\loidea) in the Ethiopian region, B. British Mus. (Natural History) 
(Entomology), 34: 1-83. 
·--. l984. Afrotropical jumping plant lice of the family Trioziclae 
(Homoptera: Psylloidea), B. British Mus. (Natural History) (Entomology), 
49: 1-102. 
----. 
1987
· A review of the Malvales-feeding psy\lid family Carsidaridae 
(Homoptera). Bull. British Mus. nat. Hist. (Ent.) 56: 87-127 
Bosl!is, D . . , . 
· and P. S. Broomfield. 1989. Ficus-feeding psyl11ds (Homoptera), with 
special reference to the Homotomidae. Bull. Br. Mus. nat. Hist. 
(Entomology) 58(2): 131-183. 
Bot1gh ' R. B. 1936. Handbook of the trees of the Northern States and Canada, 
east of h h C any Lowville, New 
t e Rocky Mountains. Romeyn B. Houg omP ' 
York. 
Bo Wden H 
' · F. and B. D. Gill. 
1988. XenocanthUS, a new genus of inquiline 
Sc-1r b (C J optera) can. J. Zool. 
c a aeidae from southeastern Venezuela 
O 
e · 
GG(9): 2071-2076. 
196 
Janson, C. 
I-I. l992. Measuring evolutionary constraints: a Markov model for 
Phylooenet' .. 
b 
1c transmons among seed dispersal syndromes. Evolution, 46(1): 
136-158. 
Jensen, D. D . . . 
· 1957. Parasites of the Psyllidae. H1lgard1a, 27: 71-99. Kr irnaszewsk. S 1
• · M. 1964. Studies on systematics of the suborder Psyllidae. Ann. 
............ _ ... 
---
Zoo!. 22(5): 1-58. [in Polish, English translation- U.S. Dept. of Commerce] 
---. 
1993a. New species of Eocene psyllids representing the tribe 
Paleops 11 'd. . ) CT Y 01 1m Becker-Migdisova (Homoptera: Aphalaridae . A A 
Biological Si1esiana, Insect Systematics and Zoogeography, 22(39): 9-18. 
, ............... _ 
-----. 
1993b. The structure of hind wings in Psylloidea (Homoptera) and its 
Possible significance in recognizing the relationships within this suborder. 
ACT A Biological Silesiana, Insect Systematics and Zoogeography, 22(39): 
57-68. 
Kwon y 
' · 1. l983. Psylloidea of Korea (Homoptera: Sternorrhyncha). Insecta 
Koreana. ser. 2. 181 pp. 
I(l'ij ss 
. n1ann, G. 1976. Manual of cultivated broad-leaved trees and shrubs. Vol. I., 
A-D T' 
· 
1mber Press, Oregon. 
liebh 
err, J. K. and A. E. Hajek. 1990. A cladistic test of the taxon cycle and taxon 
l . PUise hyPotheses. Cladistics, 6: 39-59. ltt) h . 
e, .c. L., Jr. 1971. Atlas of United States trees. Vol. 1. Conifers and important 
h-1 I . DC. w s Department of 
<rcwoocts. Misc. Publ. 1146. Washington, · .. · · 
Agricu lture, Forest Service. 
197 
• 
· Atlas of Uruted States trees. Vol. 3. Minor Western hardwoods. 
------------· 197 6 . 
Misc. Publ. 1314 
Washington, D.C.: W.S. Department of Agriculture, 
Forest s · , erv1ce. 
------------· 1977 · Atlas of United States trees. Vol. 4. Minor Eastern hardwoods. 
Washington, D.C.: W.S. Department of Agriculture, 
Misc. Publ. 1342. 
Forest Se . rv1ce. 
Loo· binova, M. . M. 1964. New and little-known psyllrds from Kazakhstan. Notes on 
system of classification of the Psylloidea (Homoptera). Trudy 
the 
Zoolo · h . · g1c eskogo mstituta. Akademiya nauk sssR 34: 52-112. [In Russian· 
English translation by Israeli programme for Scientific translation in 
Collect d · D A d NSF e Arttcles in Insect Taxonomy, 114- 177. U. S. · · an ' 
1969.] 
Mactct· Ison, W. P. 1990. A method for testing the correlated evolution of two 
binary ch d t' ·n branches of 
aracters: are gains or losses concentrate on cer a, 
a Phylog . "-,.. enetic tree. Evolution 44(3): 539-557. 
'~act ct· , 
Ison w p · f h toaenY and 
' · · and D. R Maddison. 1992. MacClade. Analysis 
O 
P Y " 
characte . d l d Massachusetts. 
M r evolution. Sinauer Assa., Joe. Sun er an , 
ally C 
' · W 18 1 proc rowa Acad. 
· 94. Hackberry Psyllidae found at AfDCS, owa. · 
Sci 189 
. 3: 1: 131-138. 
-----------. 189 , S i 1894: 2: 152-171 
,. 5. Psyllidae found at Ames. Proc. Iowa ACad. c. 
iv.tan· 
t, M. S 19 . e Biologicae, W. Junk, 
· 64· The ecology of plant galls. Monographia 
The Hague. 
198 
------
-· 
0 
uct10n to cecidology. pp. 3-7. In Sbortbouse, J. D. and O. 
-----. l99? Intr d · 
Rohfritsch [ d ] · . . 
M 
e s. B10logy of msect-mduced galls. Oxford Univ. Press. 
a ynard Smith J  ' · 1966. Sympatric speciation. Amer. Nat. JOO: 637-650. 
ayr, E. 1976. Evolution 
Cambridge, Mass. 
and the diversity of life. Harvard Univ. Press, 
Meye r, J. 1987. 
Plant galls and gall inducers. 29lpp. Gebriider Borntraeger, 
Berlin 
Micke · v1ch, M , ·, and M. S. Johnson, 1976. Congruence between morphological and 
allozym 1 . . 
e c ata m evolutionary inference and character evolutwn. Syst. 
Zoo!., 25: 260-170. 
Micke . v.1ch M 
' ·, and C. Mitter. J981. Treating polymorphic characters rn 
sys tem · 
atrcs: a phylogenetic treatment of e]ectrophoretic data. PP· 45-60 
in: Ad · 
vances m Cladistics. Proceedings of the first Meeting of the Willi 
Hen· . mg Society. V.A. funk and D.R. Brooks, eds., New York Botanical 
Gard en, Bronx NY 
tvr· ' . 
iyatak 
e y · ll'd 
' · 
19
68. Pachypsylla japonica sp. nov., a remarkable lerp-forwng psy 
1 
form Ja . . ,, . N· t !-list. 21: 5-12. 
pan (Homoptera: Psy!hdae). Bull. Osaka 1,,us. d · · 
---. 1980. Note on the genus Pachypsylla of Japan, with description of a new 
species (H k ,..,1' Nat I:{ist 33· 61-70. 
t.,i omoptera: Psyllidae). Bull. Osa a 1"us. · · · · · 
Oore I<: 
, .M. 
............ 
197 . f ·t insects 24. Results 
0. Observations on some Austrahan ores · · 
from a , . p yllidae) Australian 
study of the genus Glycaspis (Bomoptera. s · 
Zoo] . ogist 13: 343-376. 
199 
Moran, N. 1988 . 
· The evolut10n of host-plant alternation in aphids: evidence for 
izat1on as a dead end. Amer. Nat. 132: 681-706. special · · 
Morgan, F. D 1 . 
· 984. Psyl101dea of South Australia. Handbook of the Flora and 
Fauna f S 0 outh Australia. 136pp. 45 Pl. 
Moser J C ' · · 1965. The interrelationships of three gall makers and their natural 
' on ackberry ( Celtis occidentalis L.). New York State Mus. and enemies h 
Sci s . 
. erv1ce. Bull. 402, 95pp. 
Osten Sacken, B. 1861. Ueber die gallen und andere durch Insekten 
hervorg b . z . e rachte Pflanzen Deformationen in Nord-A.menca. Ent. ertung 
Stett' m. 22: 405-4?3 
Patch, E. M - . p · 1912· Notes on Psyllidae. Maine Agr. Exper. Sta., Bull. 202: 215-
234
· 
epoon, B . , . 
· S. 1927. An annotated flora of the Chicago area. Chicago Acc1d. Sc1. 
554pp. 
Pia Ochon J p ' · E_ 1873. Ulmaceae. D- C. Prodr. 17: !51-210. 
reston J . 
, - p_ Jr. 1961. North American Trees (Exclusive of Mexico and Tropical 
United s I , p . tates). The Iowa State Univ. Press, AJileS, owd 
rice, p . . . W 1980 . p ·nceton Umv. Press,
· · The evolutionary biology of parasites. n 
P· 
nnceton, NJ. 
--
----------. 1987 A . · Entornol. 16: 15-24. 
••• · daptive nature of insect galls. Environ. 
·--------. 1988 . . . flies Jn Shorthouse, J. 
· Evolution and ecology of gall-inducing swa · 
b. an i D . d Acarina induced ga!Js. 
c · Rohfritsch [eds.] Biology of insect an 
Pr·i 
c eger, N.Y. 
200 
1992. Evolution and ecology of gall-inducing sawflies. pp. 208-224. In 
Shorthou J 
se, · D. and 0. Rohfritsch [eds.] Biology of insect-induced galls. 
Oxford Univ. Press. 
Price 
'P. W G L . 
·, · · Wanng and G. W. Fernandes. 1986. Hypotheses on the 
adaptive 
nature of galls. Proc. Entomol. Soc. Wash. 88(2): 361-363. 
Price, P. W ·, G. W. Fernandes and G. L. Waring. 1987. Adaptive nature of insect 
galls. Environ. Entomol. 16: 15-24. 
Prokopy, R . . . · J. , S. R. Diehl, and S. S. Cooley. 1988. Behavioral evidence for host 
rac · es m Rhagoletis pomone/la flies. Oecologia, 76: 138-147. 
Reed, H. C . . . . 
· and R. D. Akre. 1983. Colony behavior of the obligate social pdrasite 
Ve1·p l 
. u a austriaca (Panzer) (HYmenoptera: Vespidae). Insects Sociaux. 
30(3): 259-273. 
Rehder A , . 1949. Ce/tis L. pp. 184-187. Manual of cultivated trees and shrubs, 
Hardy in North America exclusive of the subtropical and warmer 
temp r • y k erate reg10ns. The MacMillan Company, New or · 
Rich· 
drdso
1 
1986 Allozyme 
· 1, B. 1., P. R. Baverstock and M. Adams. ' '· 
electr 
1 
· d pulation studies. 
op ioresis. A handbook for animal systematics an po 
410 . pp. Academic Press 
:R.ie · lllann, J. G. 1958. Notes on the host and biology of Tetragonocep/Ja/a f/ava 
Cra f s 31(4)· 256. 
w ord (Homoptera: Psyllidae). J. Kansas Ent. oc. · 
---
----- --- 19 k genus Pachypsylla 
· 61. A study of the hackberr)' gallm• er · (Ho V · f Texas Austin. Z04PP· 
llloptera: Psyllidae ). Ph.D. dissertation. OJV. 
0 
' 
201 
41 plates. 
·····-------. 1966. Chromosome numbers in the hackberry psyllids Pachypsyila and 
Tetrago 
· nocephala (Homoptera: Psyllidae). Ann- Ent. Soc. Amer. 59(6): 
1088-1093. 
Riley, C. V . . · 1876. Gall msects. Johnsen's Universal Cycloped1a, p. 415. 
---... 
---
-----. l883. Hackberry psyllid galls. Can. Ent. 15: 157-159. 
----. lS85. The Psyllidae of the United States. Proc. Am. Ass. Adv. Sci. 
(l883) 32: 319. 
·----. 
18
90. Insects injurious to the hackberry. Chap. Xll, fifth Rept. U. S. 
Ent. Comm. pp. 601-622. 
Riley 
1 
. . ' ·, J. M. Winch, A. f. Stimson, and R. D. Pope. 1986. The assoc1at1on of 
A mp his baena alba (Reptilia: Am phis ba enia) with the leaf -cutting ant Atta 
p ialotes in Trinidad. J. Nat. Hist. 20(2): 459-470. ce / 
Ringo 
1 
. 
' · M. 
19
77. Why 300 species of Hawaiian Drosophila? Tile sexual selection 
hypothesis. Evolution, 31: 694-696. 
Rohlf, F. J. I and R. R. Sokal. 1969- Statistical Tables. W. H. Freeman ant 
Comp any, San Francisco. R..oskani 
1 
. . . female birch catins. JJ. 
· ' · C. 1979. Biosystematics of insects hvmg in 
I nqu T . to various ,,enera. Neth. 
1 me and predaceous gall midges belongmg · 
0 
J. Zoo!. 29: 283-351. 
---------- 198 . . . . f le birch catkins. IV. Egg-
. 6- B10systematics of insects ]Jvmg m ema larv eiJle and Metaciisisforster 
al parasitnides of the genrea P/atygaster Latr 
202 
d 
(Hymenoptera, Platygastridae). Tijdscbr. Entomol. 129: 125-140. 
------------. 
1992. Evolution of the gall-inducing guild. pp. 34-50. In Shorthouse, J. 
D. and O R hf. 
· o ntsch [ eds.] Biology of insect-induced galls. Oxford Univ. 
Press . 
Roskam, J. C. and G. A. van Uffelen. 1981. Biosystematics of insects Jiving in 
female birch catkins. III. Plant-insect relation between white birches, Betula 
L. Section Excelsae (Koch) and gall midges of the genus Semudobia Kieffer 
(Diptera. Cecidomyiidae). Neth. J. Zool. 31: 533-553. 
Ronq . 
Uist, F. 1994. Evolution of parasitism among closely related species: 
Phyloge t· · ·1· · · JI ne 1c relationships and the origin of rnqw rn1sm m ga wasps 
(Hymenoptera, Cynipidae). Evolution, 48(2): 241-266. 
Riibsaarn 
en, E. H. 1899. Mitteilungen iiber neue und bekannte Gallen aus 
Europa, Asien, Afrika und Amerika. Entomol. Nachr., Berlin 25: 225-282. 
Russell L . 
' · M. 1971. Notes on Euphalerus nidifex Schwarz and related nest-makmg 
New World psyllids (Homoptera: psyllidae) Florida Entomologist, 54: 3-12. 
Sargent C 
' · S. 1922. Manual of the trees of North America (2nd. ed.). Houghton 
LiffJi c S n o., Boston. 910 pp. 
chwarz, E. A. . (P ll'd e) of Australia. Proc. 1898. Notes on the lerp-rnsects sy 1 a ' 
Entorn 1 o · Soc. Washington, 4: 66-75. 
Sborth0 , . · 1 f Use, J. D. 1973. The insect community associated w1th rose gals o 
Di t Q ' t Entomol 9: 55-98. 
:p olepz'.r; polita (Cynipidae, Hymenoptera). uaes · · 
----------- 19 · b the inquilinePeric!Z:rtus 
-. 80. Modification of galls of Diplolepis polzta Y 
203 
J 
pirata. Bull. Soc. Bot. Fr. (Actual. Bot.) 127: 79-84. 
------------. l991. An unusual population of galls of Diplolepis polita (Hymenoptera: 
Cynipidae) in the Cypress Hills of Southeastern Alberta. Canadian Field-
Naturalist, 105( 4 ): 542-549. 
Shorthouse, J. D 
. and R. J. West. 1986. Role of the inquiline, Dasineura 
bafsanzicola (Diptera: Cecidomyiidae ), in the balsam fir needle gall. Proc. 
Ent. Soc. Ontario, 117: 1-7. 
Skuhrav,, M . . 
a, ·, V. Skuhravy and J. W. Brewer. 1984. Bwlogy of gall midges. pp. 
169-222. in Ananthakrishnan, T. N. [ ed.] Biology of gall insects. Oxford and 
IBH Pub. Co., New Delhi, India, 362pp. 
Smilli EL . 
' · · 1970. Biosystematics and morphology of Symphyta. II. Bwlogy of the 
gall-making nematine sawflies in the California region. Ann. Ent. Soc. 
Amer., 63: 36-51. 
Sokal R 
' · R. and F. J. Rohlf. 1981. Biometry. The principles and practice of 
statistics in biological research. 2nd ed. W. H. Freeman and Company, San 
Francisco. 
Spieg 1 h l 
e ' M. 1980. The relationship between the spatial structure of !Jranc ec 
corals and the species diversity of the inquiline fishes . Isr. 1. Zoo!. 
29( 4 ):205. 
Ste 1 U . P 
p lens, H. A. 1973. Woody Plants of the North Central Plains. The rnv. ress 
of Kansas. Manhattan. 
Swoffo d . . a parsimony ver. 3.1.1. 
r ' D. L. 1993. PAUP: Phylogenetic analysis usrn° · ' 
204 
---·· 
Illinoi s N t 1 H. . , a ura 1story Survey, Champaign. 
Swofford, D. L. 
and R. B. Selander. 1981. Biosys-1, Release 1, A computer 
program for the analysis of allelic variation in genetics. 65pp. 
Swofford, D. L. 
and Berlacher. 1987. Inferring evolutionary trees from gene 
frequency data under the principle of maximum parsimony. Syst. Zool. 36: 
293-325. 
Takagi S 
' · and Y. Miya take. 1993. SEM observations on two Jerp-forming psylloids 
(Homoptera). Insect Matsumurana, new series 49: 69-104. 
Taylor, I( L . . . . . . 
· · 1960. Add1t10nal 1nformat10n on the Australian genera of the family 
Psyliidae (Hemiptera: Homoptera). Austr. J. Zool. 8: 383-391. 
... __ _ 
-----. l962. The Australian genera Cardiaspina Crawford and Hyalinaspfr 
Taylor (Homoptera: Psyllidae). Aust. J. Zool. 10: 307-348. 
------. 1987. Ctenarytaina fongicauda sp. n. (Homoptera: psylloidea) from 
Lopfzostemon confertus (R. Brown) in Australia and California. J. Aust. ent. 
Soc. 26: 229-233. 
------. l989. Three new species of Cardiaspina Crawford (I-Iomoptera: 
PsylJoidea) from Australian eucalypts. J. Aust. ent. Soc. 28: 45-52· 
------------. 
1990. The tribe Ctenarytainini (Hemiptera: Psylloidea): a key to known 
Australian genera, with new species and two new genera. Invertebr. Taxon. 
4: 
1
't1thi1I, L. 1 of Mexico (Psy/Jidae: D. 1943. The psyllids of America nort 1 
Bomoptera). Iowa St. Acad. Sci., 17(4): 443-660. 
205 
---·· 
--------
-- 1959 Lo ,' · , 
· s Ps1l11dae del Peru Central (Insecta: Homoptera). Revista 
Peruan d E 
a e ntornologia Agricola, 2: 1-27. 
(J· Ichanco L B 
' · · 1919. A biological and systematic study of Philippine plant galls. 
Ph·1· 1 ipp. 1. Sci. 14: 527-554. 
Vondracek 
' K. 1957. Mery Psylloidea. Fauna CSR, 9: 1- 431. 
Wagner W 
' · H. 1974. Dwarf hackberry (Ulmaceae: Celtis tenuifolia) in the great 
lakes region. The Michigan Botanist, 13: 73-99. 
Walton B 
' · C. 1. 1960. Tl1e life cycle of the hackberry gall-forming, Pac/1ypsylla 
Cefta· 1 is-gemma (Homoptera: Psyllidae). Ann. Ent. Soc. Amer. 53: 265-277. 
Washb 
Urn, 1. 0. and H. V. Cornell. 1979. Chalcid parasitoid attack on a gall wasp 
Population on oak. Can. Entomol. 111: 391-400. 
Washburn J 
' · 0. and H. V. Cornell. 1981. Parasitoids, patches, and phenoJogy: 
their possible role in the local extinction of a cynipid gall wasp population. 
Ecology, 62: 1597-1607. 
Wens, B. W. 1920. Early stages in the development of certainPaclzypsyl/a galls on 
Celtis. Amer. 1. Bot. 7(7): 275-283. 
W1 · 
1Ite, T. C. R. 1970. The nymphal stage of Cardiaspina densitexta (Homoptera: 
Psyllidae) on leaves of Eucalyptus fasiculosa. Aust. Jour. Zoo/. 18: 273-293 . 
.................................... 19 [. f syllicl larvae with 72. The production of amylase in the aeces O P 
spe · 1 . d ·t ta Insect Physiol., 18: cia reference to the lerps of Cardiaspma ensz ex · 
2359-2367. 
White, 1 Jr tern) M 1982. Psylloidea (Hemiptera, -iomop ' ' 
· · and I. D. Hodkinson. 
206 
nymphal stages, Handbooks for the identification of British Insects (II) Sb: 
1-50. 
------------. 1985. 
Nymphal taxonomy and systematics of the Psylloidea 
(Bomoptera). Bull. British Mus. Nat. His. (Ent.) 50(2): 153- 301. 
Wiebes-R · .k 1J s, A. A. and J. D. Sborthouse. 1992. Ecological relationships of insects 
inhabiting cynipid galls. pp. 238-257. in J. D. Shorthouse and 0. Rohfritsch, 
eds. Biology of insect-induced galls. Oxford Univ. Press. New York. 
Wood, T. K. 
1980. Divergence in the Enchenopa binotata Say complex 
(Bomoptera: Membracidae) effect by host plant adaptation. Evolution, 34: 
147-160. 
Wood, T. K. 
and S. I. Guttman. 1982. Ecological and behavioral basis for 
reproductive isolation in the sympatric Enclzenopa binotata complex 
(Bomoptera: Membracidae). Evolution, 36: 233-242. 
Wood T . 
' . K., K. L. Olmstead, and S. I. Guttman. 1990. Insect phenology mediated 
by host plant water relations. Evolution, 44(8): 629-636. 
WoOdb · f } 
urn, T. L. and E. E. Lewis. 1973. A comparative histological study o t Je 
effects of feeding by nymphs of four psyllid species on 
Bucalypts. J. Aust. Ent. Soc. 12: 134-138. 
the leaves of 
Yano c · · , If e 
o, . K. and F. Li. 1982. Descriptions of the new genus Celttsaspts anc iv 
n . ) E taxonomia 4(3): 183-
ew species of China (Homoptera: Psyllidae · ntomo ' 
198. 
Yang, M 
. M., 1986. Reproductive Isolation and C. T. Yang and J. T. Chao. 
207 
I 
taxonomy of two Taiwanese Pauroceplzala species (Homoptera: Psy11oidea) . 
.Monograph of Taiwan Museum. Taiwan Mus. Spec. Publ. ser. No. 6, pp. 
176-203. 
208