Morphometric analyses of Palaearctic Dasineura and Jaapiella species developing in flower heads of Asteraceae, with description of four new species


Sylven, E.; Lindberg, B.

Entomologica Scandinavica 29(3): 241-265

1998


Morphometric studies of males and females of species of oligotrophine Cecidomyiidae reared from flower heads of seven genera of Asteraceae (Centaurea, Cirsium, Hieracium, Hypochoeris, Inula, Reichardia, and Saussurea) demonstrated, despite great overall morphological conformity, many instances of statistically significant differences between the species. Measurement values of structures in adult specimens were corrected for allometric deviations according to a procedure representing a further development of a method previously used for taxonomic purposes in gall midges. No galls produced by these midges have been observed. Beyond adults, also pupae and larvae were studied. Two Dasineura species and several Jaapiella species are represented. Four new species are described: Dasineura vulgatiformiae Sylven, Jaapiella hypochoeridis Sylven, Jaapiella pilosellae Sylven, and Jaapiella reichardiae Sylven.

Morphometric
analyses
of
Palaearctic
Dasineura
and
Jaapiella
species
(Diptera:
Cecidomyiidae,
Oligotrophini)
developing
in
flower
heads
of
Asteraceae,
with
description
of
four
new
species
EDVARD
SYLVEN
and
BO
LINDBERG
Ent.
scand.
Sylven,
E.
&
Lindberg,
B.:
Morphometric
analyses
of
Palaearctic
Dasineura
and
Jaapiella
species
(Diptera:
Cecidomyiidae,
Oligotrophini)
developing
in
flower
heads
of
Asteraceae,
with
description
of
four
new
species.
Ent.
scand.
29:
241-265.
Copenhagen,
Denmark.
No-
vember
1998.
ISSN
0013-8711.
Morphometric
studies
of
males
and
females
of
species
of
oligotrophine
Cecidomyiidae
reared
from
flower
heads
of
seven
genera
of Asteraceae
(Centaurea,
Cirsium,
Hieracium,
Hypochoe-
ris,
Reichardia,
and
Saussurea)
demonstrated,
despite
great
overall
morphological
con-
formity,
many
instances
of
statistically
significant
differences
between
the
species.
Measure-
ment
values
of
structures
in
adult
specimens
were
corrected
for
allometric
deviations
accord-
ing
to
a
procedure
representing
a
further
development
of
a
method
previously
used
for
taxo-
nomic
purposes
in
gall
midges.
No
galls
produced
by
these
midges
have
been
observed.
Be-
yond
adults,
also
pupae
and
larvae
were
studied.
Two
Dasineura
species
and
several
Jaapiella
species
are
represented.
Four
new
species
are
described:
Dasineura
vulgatiformiae
Sylven,
Jaapiella
hvpochoeridis
Sylven,
Jaapiella
pilosellae
Sylven,
and
Jaapiella
reichardiae
Sylven.
E.
Sylven,
Department
of
Entomology,
Swedish
Museum
of
Natural
History,
S-104
05
Stock-
holm,
Sweden.
B.
Lindberg,
Lens-Tech
AB,
Fiibodgatan
26,
S-931
56
Skelleftea,
Sweden.
Introduction
The
gall
midges
of
the
present
study
were
all
col-
lected
and
reared
from
larvae
in
the
flower
heads
of
Asteraceae
belonging
to
Centaurea,
Cirsium,
Hieracium,
Hypochoeris,
Inula,
Reichardia,
and
Saussurea.
No
galls
or
other
obvious
tissue
defor-
mations
of
the
host
plant
could
be
discerned
from
the
attacks
of
these
midges.
The
midges
retrieved
from
the
different
hosts
were
all
oligotrophines
re-
ferable
to
the
genera
Dasineura
and Jaapiella,
both
difficult
groups
characterized
by
an
extreme
morphological
conformity.
However,
on
the
basis
of
morphometric
analyses
a
good
understanding
of
the
species
taxonomy
was
obtained.
The
computer
programming
carried
out
as
a
part
of
the
present
study
constitutes
work
done
by
B.
L.
Otherwise
the
responsibility
for
the
contents
of
the
paper,
including
the
design
of
the
statistical
treat-
ment,
lies
with
E.
S.
©
Entomologica
scandinavica
(Group
6)
Material
and
methods
For
code
names
of
the
midges,
and
records
of
host
plants
and
the
geographical
origins
of
the
speci-
mens
examined,
see
Table
1.
The
specimens,
all
of
them
mounted
in
Hoyer's
medium
according
to
the
methods
described
by
Sylven
&
Neufcld
(1991),
were
examined
with
a
Zeiss
microscope
equipped
with
phase-contrast
optics.
Except
when
stated
otherwise,
specimens
have
been
deposited
in
the
gall
midge
collection
at
the
Department
of
Ento-
mology,
Swedish
Museum
of
Natural
History
(ab-
breviated
SMNH
below),
Stockholm.
In
accordance
with
methods
previously
em-
ployed
(e.g.,
Sylven
&
Lovgren
1995)
certain
adult
structures,
provided
they
were
found
to
be
in
a
horizontal
or
almost
horizontal
position
in
the
slides,
were
measured,
to
the
nearest
0.5
mm
at
the
magnifications
used
(wing
x
73
,
antenna
and
ovi-
positor
x
187,
most
other
structures
measured
x
242
Sylven,
E.
&
Lindberg,
B.
ENT.
SCAND.
VOL.
29:3
(1998)
JAC
J
PIL
SCA
HYP
INU
REI
CIR
RW
VUL
LA/
SAU
1
mm
Fig.
1.
Right
wing
of
male
Dasineura
and
Jaapiella
spp.
by
code
names.
See
'Material
and
methods'
for
explanation
of
measurements
LW
and
RW.
Table
1.
Some
relevant
data
about
the
studied
gall
midges
(listed
by
code
names).
Gall
midge
Code
name
Host
plant
Geographical
origin
Collector
JAC
Centaurea
jacea
Sweden
E.
S.
SCA
Centaurea
scabiosa
Sweden
E.
S.
CIR
Cirsium
arvense
The
Netherlands
W.
N.
CIR
C.
arvense
Germany
H.
M.
CIR
C.
arvense
Sweden
E.
S.
PIL
Hieracium
pilosella
Sweden
E.
S.
VUL
Hieracium
sp.
(group
vulgatiformia)
2
Sweden
E.
S.
HYP
Hypochoeris
radicata
Germany
H.
M.
HYP
H.
radicata
Sweden
E.
S.
INU
Inula
salicina
Sweden
E.
S.
REI
Reichardia
picroides
3
Greece
B.
S.
REI
R.
picroides
Croatia
E.
S.
SAU
Saussurea
alpina
Sweden
E.
S.
'B.
S.
(Brita
Sylven),
E.
S.
(Edvard
Sylven),
H.
M.
(Hans
Meyer),
W.
N.
(Wim
Nijveldt).
2
Identified
by
Bengt
Jonsell,
the
Bergius
Foundation,
Stockholm.
3
Identified
by
Bertil
Nordenstam,
Swedish
Museum
of
Natural
History,
Stockholm.
ENT.
SCAND.
VOL.
29:3
(1998)
Oligotrophine
gall
midges
in
flower
heads
of
Asteraceae
243
Table
2.
Statistics
of
LW
for
the
studied
gall
midges
(listed
by
code
names).
Males
Females
n
Range
(mm)
C.i.
(mm)
n
Range
(mm)
C.i.
(mm)
PIL
11
.90-1.11
.97-1.07
22
.95-1.33
1.07-1.17
HYP
8
1.06-1.34
1.09-1.15
17
1.17-1.43
1.26-1.35
CIR
13
1.16-1.60 1.34-1.50
17
1.03-1.56
1.24-1.38
INU
14
1.06-1.34
1.15-1.24
11
1.19-1.53
1.29-1.44
JAC
3
1.26-1.27
15
1.23-1.52
1.36-1.45
SCA
7
1.09-1.27
1.11-1.23
13
1.29-1.71
1.36-1.51
REI
13
.99
1
1.27
1.14-1.23
12
1.32-1.61
1.40-1.49
VUL
11
1.11-1.35
1.23-1.34
13
1.26-1.55
L40-1.50
SAU
14
1.43-1.58
1.47-1.52
15
1.60-1.77
1.64-1.69
480).
The
following
measurements
were
made:
LW,
arculus
to
wing
apex
distance
(here
consid-
ered
to
indicate
size
of
specimens)
(Fig.
1);
RW,
arculus
to
distal
point
of
R5
distance
(Fig.
1);
BW,
wing
breadth
as
indicated
by
an
imaginary
line
at
90°
to
and
crossing
the
centre
of
LW;
AL,
antennal
length;
GS,
length
of
male
gonostylus
(Fig.
3);
OV,
ovipositor
length
(anterior
limit
of
genital
chamber
to
abdominal
tip
distance)
(Fig.
4);
and
CE,
length
of
female
cercal
lamella
(distance
from
anterior
limit
of
hypoproct
to
abdominal
tip)
(Fig.
5).
In
addition,
length
of
pupal
skin,
and
in
the
last
instar
larva,
length
of
emptied
and
flattened
skin,
and
length
of
certain
setae
were
measured.
Measurement
values
referring
to
adult
struc-
tures
were
largely
treated
according
to
statistical
methods
used
by
Sylven
(in
Sylven
&
Lovgren
1995).
Thus,
the
original
values
were
assumed
to
fit
the
allometric
equation
logy
=
b
log
x
+
log
a
that
describes
a
straight
line
with
the
slope
=
b,
and
the
intercept
=
log
a.
Isometry
(proportional-
ity)
is
present
when
h
=
1
(unity)
whereas
b
>
1
and
b
<
1
indicate
positive
and
negative
allometry
respectively.
In
the
analyses
discussed
below
log
LW
corresponds
to
log
x,
and
log
RW,
log
BW,
etc.,
to
log
y.
The
following
notations
are
used:
b
=
slope
of
the
linear
regression
line.
b'
=
slope
of
the
lower
95
%
confidence
limit
of
the
slope
of
the
linear
regression
line.
b"
=
slope
of
the
higher
95
%
confidence
limit
of
the
slope
of
the
linear
regression
line.
C.i.
=
95
%
confidence
interval.
All
log
values
mentioned
below
are
'Nog
values.
Separately
for
males
and
females
of
CIR,
HYP,
etc.,
respective
linear
regression
lines
and
their
C.i.:s
referring
to
log
RW
against
log
LW,
log
BW
against
log
LW,
etc.,
were
calculated
(Tables
3-4;
Figs
6-7).
The
log
values
of
RW,
BW,
etc.,
were
corrected
for
size
of
respective
specimens
(as
indicated
by
Table
3.
Number
of
specimens
(n)
and
product-moment
correlation
coefficients
(7
-
12
)
in
the
linear
regression
analyses.
Gall
midges
listed
by
code
names.
CIR
HYP
INU
JAC
PIL
REI
SAU
SCA
VUL
n
7
-
12
n
r,
2
n
r12
n
r
12
n
r
12
n
7
-
12
n
r,
2
n
r
12
n
r
12
13
.99
8
.91
14
.99
3
11
.99
13
1.00
14
.97
7
.99
11
.
1.00
17
1.00
17
.99
11
1.00
15
.99
221
.00
12
.99
15
.96
13
1.00
13
1.00
13
.94
8
.54
14
.94
3
11
.97
13
.98
14
.73
7
.82
11
.97
17
.95
17
.96
11
.97
15
.97
22
.96
12
.78
15
.87
12
.95
13
.95
8
.98
8
.77
12
.87
2
9
.91
13
.93
13
.70
7
.81
8
.82
16
.94
14
.90
11
.90
13
.83
20
.95
11
.85
14
.35
9
.97
12
.93
15
.89
16
.82
11
.90
14
.83
22
.90
12
.66
15
.44
11
.84
13
.78
15
.45
16
.47
11
.74
14
.06
20
.58
12
.89
15
.10
11
.74
13
.45
12
.80
7
.65
13
.67
3
8
.99
12
.92
13
.31
6
.84
11
.93
log
RW
on
log
LW,
o'
log
RW
on
log
LW,
9
log
BW
on
log
LW,
o'
log
BW
on
log
LW,
9
log
AL
on
log
LW,
o'
log
AL
on
log
LW,
9
log
OV
on
log
LW,
9
log
CE
on
log
LW,
9
log
GS
on
log
LW,
o'
S
b
-1
S
b
S
b
-1
s,
S
b
-1
sb
S
b
-1
Sb
S
b
-1
sb
S
b
-1
Sb
S
b
-1
Sb
S
b
-1
Sb
1.0
1.1
1.5^
,
1.0
1.0
1.0
.1"'
1.0
-.1"'
1.1
2.0bs
.5
1.0
.9
1.0
.8 .8
1.0
1.5
1.2
1.1
1.1
1.1
1.2
1.1
1.1
1.2"
,
1.1
2.8*
1.1
3.2
1.1
5.0***
1.0
.2°
1.0
.1"
,
1.1
3.2**
1.0
1.3"
,
1.0
1.0
1.0
1.1
.9
.8
1.0
1.0
1.1
1.2
1.2
1.2
1.1
1.2 1.2
1.1
.9
1.1
1.1
.8"
,
1.6
5.5*
1.1
.2"
,
1.0
1.3
2.5*
-.5
8
.9
1.4
.4
.2
1.0
2.4
1.4
1.3
1.8
1.7
1.7
1.5
1.3
2.7*
1.1
1.0"s
1.0
.3^
,
1.1
1.2"s
.7
-2.0"
,
1.5
2.2*
1.2
1.5*
1.1
1.1"
,
1.1
.9
.8
.9
.3
1.0
.9
.9
1.5
1.3
1.1
1.2
1.0
2.1
1.5
1.4
1.3
.Ins
1.1
1.5
1.8^
,
1.2
1.4*
1.9
1.6^
,
1.0
.7^
,
.8
2
.7
.9
.9
.6
.2
.2
2.3
1.6
2.1
1.5
3.1
1.9
1.4
.9
1.1
.9
.9
-2.1*
.7
-2.1ns
.6
-1.0^s
1.1
Libs
.9
-.4"s
.6
.7
.5
.7
.4
-.4
.9
.7
1.2
1.5
1.3
1.0
1.0
1.5
1.4
1.2
1.0
1.3
1.6"
,
.8
-.9"
,
.7
-4.7***
.8
-.5^
,
.6
-1.4"
,
.9
-.7^
,
.6
-3.1*
.6
.9
.5
.5
.2
-.1
.4
.3
1.4
1.8
1.2
.8
1.5
1.3
1.3
.9
.3
-5.1***
.7
-L2"
,
-.1
-3.8**
.3
-6.6***
1.4
1.7"
,
.1
-3.0*
.6-2.7*
.3
-3.9**
.0
.2
-.7
.1
.9
-.5
.2
-.1
.6
1.2
.5
.5
1.9
.7
.9
.7
.8
.2
-10.3***
1.0
-.1"
,
.7
-3.7**
.4
-1.9"
,
.5
-3.3*
.8
-1.7"
,
-.3
.1
.8
.5
-.4
.0
.6
1.8
.4
1.2
.9
1.1
.9
1.1
S
b
-1
sb
log
RW
on
log
LW
o
b
1.0
.1"
,
b
or
b
.'
9
b"
log
BW
on
log
LW
b
1.1
.6"
,
b
.8
b"
1.3
9
b
1.1
.7"
,
9b"
.9
9b"
1.3
log
AL
on
log
LW
o
b
1.3
2.6*
b
1.0
b'
o
1.6
9b
.9
-1.3"
,
9
b"
.7
9
b'"
1.1
log
OV
on
log
LW
9
b
.6
-4.1**
9
b
.4
9
b"
.8
log
CE
on
log
LW
9
b
.3
-3.9**
9
b'
-.1
9
b
"
.7
log
GS
on
log
LW
o
b
.6
-2.3*
b
.3
o
b"
1.0
244
Sylven,
E.
&
Lindberg,
B.
ENT.
SCAND.
VOL.
29:3
(1998)
Table
4.
Regression
statistics
to
test
to
what
extent
positive
or
negative
allometry
is
present
between
certain
variables.
Gall
midges
listed
by
code
names.
For
number
of
specimens
see
Table
3.
S
=
slope;
b
=
slope
of
the
linear
regression
line;
b'
and
b"
=
lower
and
higher
95
%
confidence
limit,
respectively,
to
slope
of
the
linear
regression
line;
(b-1)
/
S
b
indicates
'significance'
level
of
deviation
of
slope
of
the
linear
regression
line
from
proportionality
(see
Sokal
&
Rohlf
1995:
471-472);
ns
=
not
significant;
*
P
<
0.05;
**
P
<
0.01;
***
P
<
0.001.
CIR
HYP
INU
JAC
PIL
REI
SAU
SCA
VUL
LW;
see
above)
by
conversion
according
to
b,
b',
or
b"
into
log
values
corresponding
to
log
LW
=
3.0702
(=
log
of
1200
pm)
in
males,
and
-
if
other-
wise
not
stated
-
to
log
LW
=
3.1461
(=
log
of
1400
pm)
in
females.
On
average,
males
are
small-
er
than
females
(Table
2),
and
this
is
the
reason
why
for
the
conversion
in
analyses
of
males
a
low-
er
log
LW
value
was
chosen
than
in
analyses
of
fe-
males.
RW„
,
BW,
,
etc.,
refer
to
means
of
log
values
converted
according
to
b,
and
expressed
as
100
antilog/1200
and
100
antilog/1400
for
males
and
females
respectively
(i.e.
as
a
percentage
of
LW).
RW
ci
,
BW,
I
,
etc.,
and
RW
Ch
,
BW
Ch
,
etc.,
refer
to
means
of
log
values
converted
according
to
b'and
b"respectively,
and
expressed
as
100
antilog/1200
(males)
or
100
antilog/1400
(females).
For
the
added
span
of
the
C.i.:s
of
RW
ei
and
RW
Ch
respec-
tively,
of
BW„
and
BW
Ch
respectively,
etc.
(i.e.
for
the
interval
between
the
highest
and
the
lowest
value
in
respective
pairs
of
C.i.:s),
the
term
toler-
ance
interval
(=
T.i.)
is
here
introduced
(Fig.
8).
ENT.
SCAND.
VOL.
29:3
(1998)
Oligotrophine
gall
midges
in
flower
heads
of
Asteraceae
245
With
T.i.
of
RW,
is
meant
the
T.i.
calculated
on
the
basis
of
the
C.i:s
of
RW
ei
and
RW
ch
,
with
T.i.
of
BW
e
the
T.i.
calculated
on
the
basis
of
the
C.i:s
of
BIM,
'
and
BW
Ch
,
etc.
Diagrams
illustrating
relationships
between
cor-
rected
values
of
one
variable
(e.g.
OV,)
on
the
ver-
tical
axis,
and
of
another
variable
(e.g.
CE
e
)
on
the
horizontal
axis
were
produced,
separately
for
males
and
females.
For
each
kind
of
midge
(CIR,
HYP,
etc.),
on
the
basis
of
the
values
converted
according
to
respective
b:s,
a
cross
was
marked,
with
the
centre
indicating
respective
means,
and
the
arms
of
the
cross
respective
C.i.:s.
The
corre-
sponding
crosses
for
values
calculated
on
the
basis
of
respective
h':s
and
b":s
were
not
inserted
in
the
diagrams,
but
instead,
for
each
kind
of
midge,
a
rectangle
limiting
the
area
influenced
by
these
crosses
was
defined.
These
rectangles
correspond
to
respective
T.i.:s
(Figs
11-13).
A
more
simplified
method,
explained
in
Fig.
20,
was
used
in
the
treatment
of
length
measurements
of
larval
setae.
Results
of
morphometric
analyses
For
the
general
morphology
of
wings,
male
geni-
talia,
male
gonostylus,
ovipositor,
and
female
cer-
eal
lamella,
see
Figs
1-5.
These
drawings
show
many
examples
of
the
structural
conformity
of
the
midges
included
in
this
study.
It
is
true
that
certain
differences
are
indicated
in
these
drawings
but
they
refer
to
dissimilarities
in
proportions.
There-
fore,
in
order
to
make
a
good
evaluation
of
the
tax-
onomic
status
of
the
midges
mentioned
in
Table
1,
it
was
considered
desirable
to
perform
morpho-
metric
analyses,
and
thereby
take
into
account
al-
lometric conditions.
The
results
of
the
morphometric
analyses
are
summarized
in
Figs
9-10.
There
is
in
most
cases
a
good
correspondence
between
the
values
based
on
ratios
of
original
arithmetic
figures
(RW/LW,
BW/LW,
etc.)
on
one hand
and
ratios
based
on
log
figures
converted
according
to
the
slope
of
the
lin-
ear
regression
line
(RW
e
,
BW
e
,
etc.)
on
the
other
hand.
But
there
are
several
striking
exceptions.
Some
items
contributing
to
such
discrepancies
are
as
follows:
(1)
Significant
allometric
deviations,
as
found
in
certain
analyses
referring
to
PIL
(Figs
9B;
10C,
D;
Table
4),
and
in
an
analysis
referring
to
SAU
(Fig.
10D;
Table
4).
(2)
Comprehensive
interpolations
at
the
con-
verting
operations.
On
an
average
the
size
of
the
adult
specimens
examined,
as
indicated
by
LW,
is
smaller
in
PIL
and larger
in
SAU
than
in
the
re-
maining
midges
here
analysed
(Table
2).
This
means
that
the
log
values
at
the
conversion
opera-
tions
were
moved,
on
an
average,
a
large
distance,
corresponding
to
a
larger
part
of
the
scale
of
the
x-
axis,
in
PIL
and
SAU
than
in
the
other
midges
in-
vestigated.
(3)
Weak
correlation,
such
as
found
in
most
re-
gression
analyses
referring
to
SAU
(Table
3).
In
particular,
weak
correlation
combined
with
comprehensive
interpolations
has
resulted
in
ex-
traordinarily
extensive
T.i.:s,
indicating
in
such
cases
a
high
degree
of
uncertainty
of
respective
means
based
on
log
figures
converted
according
to
the
linear
regression
line.
This
is
also
exemplified
by
the
outcome of
many
analyses
based
on
the
val-
ues
referring
to
SAU.
Figs
9-13
and
Table
5,
the
latter
showing
the
re-
sults
of
significance
tests
according
to
the
Student-
Newman-Keuls
test
(see
Sokal
&
Rohlf
1969:
242-245),
give
clear
evidence
to
show
that
at
least
eight
of
the
nine
kinds
of
midges
(CIR,
HYP,
etc.)
here
studied
can
be
distinguished
from
each
other
on
the
basis
of
morphometric
characters.
The
diagrams,
Figs
11-13,
give
clear
indications
of
the
applicability
for
taxonomic
purposes
of
cer-
tain
of
the
characters
analysed.
In
Fig.!
1,
based
on
females
and
dealing
with
BW
e
against
RW,
,
it
can
be
seen,
that
SAU,
despite
very
large
T.i.:s,
is
clearly
distinguished
from
REI,
PIL,
and
HYP.
Al-
so,
as
shown
in
Fig.
13,
referring
to
females
and
dealing
with
OV,
against
RW,
,
SAU
is
clearly
dis-
tinguished
from
JAC
and
SCA.
Several
other
con-
clusions
can
be
drawn
from
Figs
11-13.
For
exam-
ple,
PIL
and
HYP,
according
to
Fig.
11,
cannot
be
distinguished
on
the
basis
of
BW
c
or
RW,
but,
as
shown
in
Figs
12-13,
differ
from
each
other
re-
garding
OV,.
A
discrepancy
from
what
has
been
said
above
is
indicated
in
Fig.
14.
The
diagram
of
that
figure,
re-
ferring
to
females,
is
based
on
log
values
convert-
ed
into
values
corresponding
to
log
LW
=
3.2041
(=
log
of
1600
pm,
as
shown
in
Table
2
a
value
closely
approaching
the
mean
of
LW
of
the
fe-
males
of
SAU
investigated).
Otherwise
the
meth-
ods
used
in
constructing
Fig.
14
agree
with
those
of
Fig.
13.
It
will
be
seen
in
Fig.
14
that
respective
T.i.:s
of
SAU
are
in
this
case
clearly
different
from
most
of
those
referring
to
females
of
the
other
midges
included
in
the
present
study.
246
Sylven,
E.
&
Lindberg,
B.
ENT.
SCAND.
VOL.
29:3
(1998)
Table
5.
Distribution
of
significant
(P
<
0.05
or
better)
differences
as
indicated
by
Student-Newman-Keuls
test
(see
Sokal
&
Rohlf
1969:
239,
and
Rohlf
&
Sokal
1969:
200ff.)
between
means
of
figures
transformed
according
to
the
slope
of
the
linear
regression
line
(RW
c
,
BW,,,
etc.)
or
according
to
the
lower
(RW,,,
BW,,,
etc.)
or
higher
(RW
Ch
,
BW„,
etc.)
95%
confidence
limit
to
slope
of
this
line.
S
=
difference
significant
in
all
combinations
of
means
between
the
two
kinds
of
midges
indicated.
=
difference
not
significant
in
one
or
more
combinations
of
means
between
the
two
kinds
of
midges
indicated.
RW
cr
RW
9
BWo
BW
9
AL
o'
AL
9
OV
9
CE
9
GS
0'
CIR
HYP
"
INU
"
JAC
"
PIL
"
REI
"
SAU
"
SCA
"
VUL
HYP
INU
"
JAC
"
PIL
"
REI
"
SAU
"
SCA
"
V
UL
INU
JAC
"
PIL
"
REI
"
SAU
"
SCA
"
VUL
JAC
PIL
"
REI
"
SAU
"
SCA
"
VUL
PIL
REI
"
SAU
"
SCA
"
V
UL
REI
SAU
"
SCA
"
VUL
SAU
SCA
"
VUL
SCA
VUL
S S
S
S
S
*
S
S
S
S
S
S
S
S S
S
*
S S
S
S
S
S
S
S
S
S S
S
S
*
S
S
*
S
S
S
*
S
S S
S
S
S
S
S
S
S S
S
S
S
S S
S
S
S
S
S
S
S
S
S
S
S
S S
S
S
S
S S
S
S
S
-
-
S
S
S
S
S
S S
S S S
*
S
S
S
S
S
S
S S
S
*
S
*
S S
*
S S
*
S S
S
S
S S
S
S
S
S
S S
*
S
S
S
S
S
S
S
*
*
S S
S
S
S
S
S
S
*
S
S
S
S
S
*
S
S
S
S
*
S
S
S
S
S
S
*
S
S
S
*
S
*
*
*
Figures
not
sufficient
for
statistical
comparison
because
of
low
number
of
specimens
(JAC
cr
only
3
specimens)
or
because
of
extraordinarily
large
tolerance
interval
(SAU).
Generic
assignments
Gagne
et
al.
(1997)
proposed
that
the
unrecog-
nized
taxon
Dasineura
obscura
Rondani
should
be
replaced
by
D.
sisymbrii
(Schrank)
as
type
species
of
the
genus
Dasineura
Rondani,
1840.
For
a
de-
tailed
redescription
of
D.
sisymbrii
see
Sylven
in
Sylven
&
Tastas-Duque
(1993:
292).
As
presently
defined,
Dasineura
is
a
genus
without
no
obvious
synapomorphies.
Of
the
midges
analyzed
below
two
species,
corresponding
to
VUL
and
SAU
(Ta-
ble
1),
are
referred
to
Dasineura,
the
remaining
species
to
Jaapiella Rfibsaamen,
1915.
Species
of
Jaapiella
possess
a
swelling
on
the
inner
distal
half
of
the
gonocoxite;
no
such
swelling
is
present
in
Dasineura
(Fig.
2).
This
is
the
only
known
fea-
ture
for
distinguishing
these
two
genera.
Thus,
it
/VUL
100
µm
SAU
ENT.
SCAND.
VOL.
29:3
(1998)
Oligotrophine
gall
midges
in
flower
heads
of
Asteraceae
247
)
PTL
/
JAC
i
SCA
I
CIR
Fig.
2.
Male
terminalia,
ventral
view,
and
gonocoxite,
dorsal
view
of
Dasineura
and
Jaapiella
spp.
by
code
names.
has
been
called
in
question
(Mohn
1955:
205;
arate
from
Das•ineura.
However,
more
taxonomic
Sylven
1975:
90;
Roskam
&
Kofman
1987:
59),
work
is
required
to
find
out
the
proper
status
of
whether
Jaapiella
is
to
be
retained
as
a
genus
sep-
Jaapiella.
248
Sylven,
E.
&
Lindberg,
B.
ENT.
SCAND.
VOL.
29:3
(1998)
JAC
-I
(7.4-
4
--
SCA
G'
.
INU
PIL
HYP
.
'--
,,)
11E1
CIR
VUL
_
GS
SAU
54
gm
Fig.
3.
Gonostylus,
dorsal
(left)
and
ventral
(right)
views,
of
Dasineura
and
Jaapiella
spp.
by
code
names.
See
'Ma-
terial
and
methods'
for
explanation
of
measurement
GS.
General
characteristics
of
the
studied
midges
Description.
Adult.
Antenna:
each
flagellomere,
except
for
the
distal
one,
consisting
of
a
node
and
a
neck,
the
latter
being
much
longer
in
males
than
in
females
(Fig.
15);
circumfila
appressed
against
the
node
in
both
sexes;
flagellomeres
1-11
not
separat-
ed
from
each
other.
Palpifer
present.
Palpus
4-seg-
mented.
Each
tarsus
(Fig.
15)
with
toothed
claws;
empodium
about
as
long
as
claws.
Wing
(Fig.
1):
R,
joining
C
anterior
to
apex;
border
of
wing
just
beyond
insertion
of
R
s
with
a
naked
patch
devoid
of
thickening.
Abdomen:
ground
colour
orange
to
red
orange;
tergites
(Fig.
15)
1-VI
each
with
a
cau-
dal
row
of
setae;
sternites
II-VII
each
with
two
transverse
bands
of
setae
(more
or
less
mixed
with
narrow,
pointed
scales);
male
terminalia
(Fig.
2)
with
cereal
lamella
and
hypoproct
bibbed,
and
with
each
of
the
two
mediobasal
lobes
sheathing
aedeagus;
microtrichial
cover
of
gonostylus
ex-
tended
dorsally
less
than
halfway
to
the
tip
(Fig.
3);
female
terminalia
with
cerci
fused
into
a
single
lamella.
Pupa
(Fig.
16).
Basal
part
of
antenna'
sheath
an-
teriorly
with
a
tiny
tooth.
Face
without
protuber-
ances.
Prothoracic
horn
elongated,
tapering
from
base
to
tip.
Abdomen
with
micro-
and
macro-
spines,
the
latter
exclusively
on
a
well-defined
area
dorsally
on
each
of
uromeres
Each
of
the
interior
dorsal
papillae
of
uromere
VII
without
se-
ta.
Last
instar
larva
(Figs
17-19).
Spatula
present.
Number
and
distribution
of
papillae
and
setae
on
thorax
and
uromeres
I-VIII
as
shown
in
Sylven
(1975:
fig.
5).
Biology.
Larvae
feed,
normally
gregariously,
in
flower
heads
of
Asteraceae
(but
as
far
as
known
not
inside
flowers
or
fruits)
of
their
respective
host
plant(s).
They
cause
little
if
any
deformation
of
the
flower
heads.
Pupation
as
a
rule
takes
place
out-
side
the
flower
head,
in
litter
or
soil.
Treatment
of
species
Dasineura
vulgatiformiae
Sylven,
sp.
n.
(Code
name:
VUL)
Identification.
The
only
name
that
might
pos-
sibly
be
available
for
the present
species
is
Ceci-
domyia
compositarum
Kieffer.
However,
in
view
of
the
narrow
host
range
encountered
for
this
group
of
gall
midges
there
is
reason
to
believe
that
Kieffer
(1888)
based
his
original
description
of
C.
ENT.
SCAND.
VOL.
29:3
(1998)
Oligotrophine
gall
midges
in
flower
heads
of
Asteraceae
249
HYP
JAC
JAC
SCA
I
ti
PIL
1
I
CIR
SAU
Fig.
4.
Ovipositor
in
dorsal view
of
Dasineura
and
Jaapiella
spp.
by
code
names.
See
'Material
and
methods'
for
ex-
planation
of
measurement
OV.
250
Sylven,
E.
&
Lindberg,
B.
ENT.
SCAND.
VOL.
29:3
(1998)
JAC
SCA
INU
CIR
SAID
4
REI
PIL
VUL
HYP
1
U
g.
Fig.
5.
Female
cereal
lamella
in
dorsal
view
of
Dasineura
and
Jaapiella
spp.
by
code
names.
Hypoproct,
situated
ven-
trally,
indicated
by
dotted
line.
See
'Material
and
methods'
for
explanation
of
measurement
CE.
compositarum
(now
assigned
to
Jaapiella,
see
Skuhrava
1986)
on
a
mixture
of
species,
viz.
one
with
larvae
developing
in
flower
heads
of
Hypo-
choeris,
a
second
with
larvae
developing
in
flower
heads
of
Hieracium
pilosella,
and
probably
a
third
species
with
larvae
developing
in
flower
heads
of
another
Hieracium
species.
As
no
type
material
seems
to
exist,
C.
compositarum
is
here
regarded
as
a
nomen
dubium.
Description.
Adult.
No.
of
flagellomeres
14-15
in
both
males
(n
=
11)
and
females
(n
=
10).
Fe-
male
cereal
lamella
in
dorsal
aspect
narrowing
more
strongly
towards
the
tip
than
in
all
other
spe-
cies
examined
(Fig.
5).
From
the
Jaapiella
species
it
is
distinguished,
in
addition
to
the
deviating
shape
of
the
gonocoxite
(see above
under
'Generic
assignments'),
by
differences
related,
e.g.,
to
OV
(Figs
4,
10C;
Table
5)
and/or
CE
(Figs
5,
10D;
Ta-
ble
5).
See
also
Figs
11-12
for
morphometric
dif-
ferences
between
this
and
the
other
species.
Pupa.
Length
of
flattened
skin
1.6-2.1
and
1.8-
2.1
mm
in
males
(n
=
4)
and
females
(n
=
7)
re-
spectively.
Last
instar
larva.
Ground
colour
orange.
Length
of
flattened
skin
2.3-2.9
mm
(n
=
12).
Spatula
in
front
with
a
pair
of
pointed
lobes,
separated
by
deep
incision
(Fig.
17).
Uromeres
I-VII
each
with
the
anterior
ventral
papillae
situated
at
the
rear
of
an
area
densely
covered
with
strikingly
small,
pointed
verrucae,
arranged
in
more
distinct
trans-
verse
rows
than
in
the
other
species
(Fig.
18).
Setal
length,
in
(7c
,
c
of
length
of
flattened
skin,
on
outer-
most
dorsal
papilla
on
prothorax
clearly
shorter
than
in
most
of
the
other
species
(Fig.
20).
Biology.
Host
plant
Hieracium
sp.
(group
vulgat-
iformia).
All
adults
obtained
emerged
after
hiber-
nation.
Type
material.
Holotype:
cf
reared
from
larva
collected
in
flower
head
of
Hieracium
sp.
(group
yulgatiformia)
in
Sweden:
Oland,
Edda
crown
forest
area,
Fiskasen,
9.vii.1985,
E.
Sylyen,
slide
5762
(SMNH).
Paratypes:
from
larvae
collected
on
same
date
and
host
plant
and
in
same
locality
as
the
holotype,
8
0',
9
9,
I I
pupal
skins,
and
12
skins
of
last
instar
larvae,
and
from
larvae
col-
lected
on
13.vii.1987
on
same
host
plant
and
in
same
lo-
cality
as
above,
2
0',
4
9
and
3
pupal
skins,
E.
Sylyen
(SMNH).
Dasineura
saussureae
Fedotova,
1996
(Code
name:
SAU)
Identification.
SAU,
the
midge
reared
from
lar-
vae
found
in
flower
heads
of
Saussurea
alpina
is
here
considered
conspecific
with
Dasineura
saus-
-
1
4-
CIS
2
HY
P
p
=
INU
2
JAC
0
a
PIL
.2
R
El
c
=
SAU
SCA
2
VUL
ENT.
SCAND.
VOL.
29:3
(1998)
Oligotrophine
gall
midges
in
flower
heads
of
Asteraceae
251
3.2
3,15
3.1
3.05
3
2.95
29
2.95
3
3.05
3.1
3.15
3.2
3.25
3.3
Lag
LW
Fig.
6.
Relationship
in
females
between
log
RW
and
log
LW.
Each
symbol
one
specimen;
lines
=
linear
regression
lines.
sureae,
a
species
described
by
Fedotova
(1996)
from
Kazakhstan
with
larvae
developing
in
flower
heads
of
Saussurea
latifolia.
The
wing
and
male
genitalia
of
D.
saussureae
illustrated
by
Fedotova
(1996:
pl.
1)
are
exceedingly
similar
to
those
of
SAU
(see
Figs
1-2).
Another
species,
also
from
Kazakhstan,
Dasi-
neura
bursakovi
Fedotova,
1996
reared
from
lar-
vae
found
in
flower
heads
of
Saussurea
elegans,
is
characterized
by
RW
ending
much
closer
to
the
wing
apex
than
in
D.
saussureae
(cp.
Fedotova
1996:
pls
1-2
and
Fig.
1
of
the
present
paper).
Description.
Adult.
No.
of
flagellomeres
15-17
in
both
males
(n
=
14)
and
females
(n
=
18).
Fe-
male
cereal
lamella
in
dorsal
aspect
less
strongly
narrowing
towards
the
tip
than
in
D.
vulgatiformi-
ae
(Fig.
5).
From
the
Jaapiella
species
D.
saussu-
reae
is
distinguished,
in
addition
to
the
deviating
shape
of
the
gonocoxite
(see
above
under
'Generic
assignments'),
by
differences
related
to
OV
(Figs
4,
10C;
Table
5)
and/or
CE
(Figs
5,
10D;
Table
5),
and
from
PIL
by
differences
related
to
RW
(Figs
1,
9A,
B;
Table
5).
See
also
Figs
13-14
for
morpho-
metric
differences
between
this
and
the
other
spe-
cies.
Pupa. Length
of
flattened
female
skin
1.7
mm
(n
=
1).
Last
instar
larva.
Ground
colour
orange
(gut
contents
of
greenish
colour).
Length
of
flattened
skin
2.0-3.1
mm
(n
=
9).
Setal
length,
in
'Yoe
of
length
of
flattened
skin,
on
outermost
dorsal
papil-
la
on
prothorax
longer
than
in
VUL
but
shorter
than
in
several
of
the
other
midges
examined
(Fig.
20).
Biology.
Host
plants
Saussurea
latifolia
and
S.
alpina.
From
sample
gathered
in
1976
one
speci-
men
emerged
indoors
in
September
same
year,
while
two
specimens
emerged
after
having
hiber-
nated
twice;
remaining
adults
emerged
after
hav-
ing
hibernated
once.
Material
examined.
From
larvae
collected
in
flower
heads
of
Saussurea
alpina
in
1976
15
0',
11
y,
10
pupal
skins,
and
10
skins
of
last
instar
larvae,
SWEDEN:
Tome
lappmark,
Njulla,
700-800
m
above
sea
level,
E.
Sylven;
from
larvae
collected
from
the
same
host
plant
in
1976,
8
y,
SWEDEN:
Tome
lappmark,
Jicpren,
E.
Sylven.
+*
CIA
2
HYP
z
1NU
0
=
JAC
2
PIL
A
2
REI
a
:
SAU
=
SCA
=
VUL
3.15
3.1
3.05
0
3
0
I.
---
-
a
2.95
2.9
2.85-
0
0
b
To
ler
anc
e
in
te
rv
a
l
252
Sylven,
E.
&
Lindberg,
B.
ENT.
SCAND.
VOL.
29:3
(1998)
2.95
3
3.05
3.1
3.15
3.2
3.25
3.3
Log
LW
Fig.
7.
Relationship
between
log
OV
and
log
LW.
Each
symbol
one
specimen;
continuous
lines
=
linear
regression
lines;
broken
lines
=
95%
confidence
limits
to
slope
of
respective
regression
lines.
Jaapiella
hypochoeridis
Sylven,
sp.
n.
(Code
name:
HYP)
Description.
Adult.
No.
of
flagellomeres
12
and
11-12
in
males
(n
=
9)
and
females
(n
=
20)
re-
Mean
and
its
95
confidence
interval
of
values
converted
according
to
Fig.
8.
Schematic
presentation
of
the
concept
tolerance
interval
(=
T.i.).
spectively.
Female
cereal
lamella
in
dorsal
aspect
less
narrowing
towards
tip
than
in
REI
(Fig.
5),
and
proportionately
much
broader
than
in
the
re-
maining
species
(Fig.
5).
In
addition,
it
is
morpho-
metrically
distinguished
from
the
other
species
by
ENT.
SCAND.
VOL.
29:3
(1998)
Oligotrophine
gall
midges
in
flower
heads
of
Asteraceae
253
A
YUL
SAU
RE1
INU
SCA
B
HYP
.
•=2.61
,
:cs=rr
i
PtL
85
90
95
C
a
Pip
PIL
HYP
VUL
B
VUL
4=11
e=p
SAU
REV
C1R
4
INU
•••
SCA
4
JAC
HYP
4====A
85
90
95
REI
P
IL
HYP
VUL
SAU
S
IR
SCA
Y.¢1.1,122=
JAC
4
CIF
tNLI
4
SAU
"I.
35
40
45
50
35
40
45
50
Fig.
9.
Morphometric
data.
(A)
Males:
C.i.
of
100
RW/LW
(white
bars),
C.i.
of
RW
:
(black
bars),
and
T..i.
of
RW
e
(dotted
bars).
(B)
Females:
otherwise
as
in
A.
(C)
Males:
C.i.
of
100
BW/LW
(white
bars),
C.i.
of
BW,
(black
bars),
and
T.i.
of
BW,
(dotted
bars).
(D)
Females:
otherwise
as
in
C.
4
indicates
significant
positive
allometry.
For
further
explanation
see
under
the
heading
'Material
and
methods',
and
Table
5.
differences
relating
to
OV
(Figs
4,
10C;
Table
5)
or
RW
(Figs
1,
9B;
Table
5).
Pupa.
Length
of
flattened
female
skin
1.8-2.2
mm
(n
=
5).
Last
instar
larva.
Ground
colour
orange.
Length
of
flattened
skin
2.3-3.1
mm
(n
=
13).
Setal
length,
in
%,5
of
length
of
flattened
skin,
on
outermost
dor-
sal
papilla
on
prothorax
longer
than
in
VUL,
INU
and
SAU
(Fig.
20).
Biology.
Host
plant
Hypochoeris
radicata.
More
than
one
generation
a
year.
Type
material.
Holotype:
9
reared
from
larva
collected
in
flower
head
of
Hypochoeris
radicata
in
GERMANY:
Husum,
25.viii.
1971,
H.
Meyer,
slide
2979
(SMNH).
Paratypes:
from
larvae
collected
on
same
date
and
host
plant
and
in
same
locality
as
the
holotype,
8
0'
and
9
9,
H.
Meyer
(SMNH).
Additional
material.
From
larvae
collected
in
1971,
host
plant
as
above,
2
9,
and
9
skins
of
last
instar
larvae,
GERMANY:
Schleswig,
Moor
ldstedt,
H.
Meyer;
from
larvae
collected
in
1987,
host
plant
as
above,
5
9
and
5
pupal
skins,
SWEDEN:
Oland,
Byxelkrok,
E.
Sylven.
Jaapiella
pilosellae
Sylven,
sp.
n.
(Code
name:
PIL)
Description.
Adult.
No.
of
flagellomeres
10-12
in
both
males
(n
=
12)
and
females
(n
=
22).
This
species
is
distinguished
from
the
other
species
by
differences
relating
to
OV
(Figs
4,
10C;
Table
5)
or
RW
(Figs
1,
9A,
B;
Table
5).
See
also
Figs
12-
PIL
H
YP
VUL
CIR
SAU
JAC
SCA
REI
ENT.
SCAND.
VOL.
29:3
(1998)
Oligotrophine
gall
midges
in
flower
heads
of
Asteraceae
255
50
45
35
y
U
co
40
85
86
87
88
89
90
91
92
93
94
95
RW
c
Fig.
11.
Relationships
in
females
of
PIL,
REI,
etc.,
between
BW,
and
RW,.
Crosses
=
means
and
their
C.i.:s
of
BW,
(vertical
axis)
and
of
RW,
(horizontal
axis).
Rectangles
=
tolerance
intervals
of
BW,
(vertical
axis)
and of
RW,
(hori-
zontal
axis).
13
for
morphometric
differences
between
this
and
the
other
species.
Pupa.
Length
of
flattened
skin
1.4-1.9
and
1.7-
2.1
mm
in
males
(n
=10)
and
females
(n
=
11)
re-
spectively.
Last
instar
larva.
Ground
colour
cadmium
orange.
Length
of
flattened
skin
2.2-2.7
mm
(n
=
15).
Setal
length,
in
Voc
of
length
of
flattened
skin,
on
outermost
dorsal
papilla
on
prothorax
longer
than
in
VUL,
INU
and
SAU
(Fig.
20).
Biology.
Host
plant
Hieracium
pilosella.
Type
material.
Holotype:
9
reared
from
larva
collected
in
a
flower
head
of
Hieracium
pilosella,
SWEDEN,
Uppland,
Danderyd,
19.vi.
1986,
E.
Sylven,
slide
6117
(SMNH).
Paratypes:
from
larvae
collected
on
same
date
and
host
plant
and
in
same
locality
as
the
holotype,
4
or,
5
9,
and
4
pupal
skins,
E.
SyIv&
(SMNH).
Additional
material.
From
larva
collected
in
vi.1978,
host
plant
as
above,
1
9,
SWEDEN:
Oland,
Drostorp,
E.
Sylven;
from
larvae
collected
in
vii.1987,
host
plant
as
above,
8
or,
7
9,
20
pupal
skins,
and
6
skins
of
last
instar
larvae,
SWEDEN:
Oland,
Byxelkrok,
E.
Sylven;
from
larvae
collected
in
vii.1963-1981,
host
plant
as
above,
1
ow,
9
9,
and
3
skins
of
last
instar
larvae,
SWEDEN:
Oland,
Both
crown
forest
area,
E.
Sylven;
from
larvae
collected
in
vii.1967,
host
plant
as
above,
7
skins
of
last
instar
larvae,
SWEDEN:
Gotland,
Hoburgen,
E.
Sylven.
Jaapiella
reichardiae
Sylven,
sp.
n.
(Code
name:
REI)
Description.
Adult.
No.
of
flagellomeres
11-12
and
12
in
males
(n
=
13)
and
females
(n
=
13)
re-
spectively.
Female
cereal
lamella
in
dorsal
aspect
proportionately
broader
at
midlength
than
in
the
other
species
apart
from
HYP
(Fig.
5).
In
addition,
it
is
distinguished
from
the
other
species
by
differ-
ences
related,
e.g.,
to
CE
(Figs
5,
10D;
Table
5).
See
also
Figs
11-12
for
morphometric
differences
between
this
and
the
other
species.
Pupa. Length
of
flattened
male
skin
(n
=
1)
1.5
mm
and
of
ditto
female
skin
(n
=
3)
1.9
mm.
Last
instar
larva.
Ground
colour
orange.
Length
of
flattened
skin
2.2.-3.2
mm
(n
=
10).
Setal
length,
in
%
of
length
of
flattened
skin,
on
outer-
most
dorsal
papilla
on
prothorax
clearly
longer
than
in
most
of
the
other
species
(Fig.
20).
256
Sylven,
E.
&
Lindberg,
B.
ENT.
SCAND.
VOL.
29:3
(1998)
95
Fig.
12.
Relationships
in
females
of
JAC,
SCA,
etc.,
between
OV,
and
CE..
Crosses
90
=
means
and
their
C.i.:s
of
OV,
(vertical
axis)
and
of
CE,
(horizontal
axis).
Rec-
tangles
=
tolerance
intervals
of
OV,
(ver-
JAC
SCA
tical
axis)
and
of
CE,
(horizontal
axis).
85
Note:
SAU
omitted.
80
75
>t'
0
HYP
NL
CI
r
70
65
PIL
REI
60
VUL
55
50
5
5.5
6
6.5
7
7.5
8
CE.
Biology.
Host
plant
Reichardia
picroides.
Pupa-
tion
may
occasionally
take
place
in
the
flower
head.
The
rearing
results
indicate
that
there
is
more
than
one
generation
a
year.
Type
material.
Holotype:
9
reared
from
larva
collected
in
a
flower
head
of
Reichardia
picroides,
Greece:
Delft,
4.xi.1982,
Brita
Sylven,
slide
3333
(SMNH).
Paratypes:
from
larvae
collected
on
same
date
and
host
plant
and
in
same
locality
as
the
holotype,
13
0',
9
9,
and
6
last
instar
larvae,
Brita
Sylven
(SMNH).
Additional
material.
From
larvae
or
pupae
collected
in
iv.
1983,
host
plant
as
above,
3
9,
6
pupal
skins,
and
4
last
instar
larvae,
Croatia:
Dalmatia,
Medena
or
(the
last
instar
larvae)
Split,
E.
Sylven.
Jaapiella
inulicola
Fedotova,
1993
(Code
name:
INU)
Identification.
The
gallmidge
(INU)
found
in
flower
heads
of
Inula
salicina
in
Sweden
is tenta-
tively
identified
with
Jaapiella
inulicola,
a
species
described
by
Fedotova
(1993)
on
the
basis
of
spec-
imens
reared
from
the
same
host
plant
in
Kazakh-
stan.
Unfortunately,
the
short
original
description
of
J.
inulicola
contains
no
information
about
AL,
OV,
or
about
the
morphology
of
the
larva,
but
the
illustrated
wing
and
male
genitalia
(Fedotova
1993:
pl.
7)
reveal
no
obvious
differences
from
the
corresponding
structures
in
INU
(see
Figs
1-2).
Description.
Adult.
No.
of
flagellomeres
12-14
in
both
males
(n
=
14)
and
females
(n
=
12).
It
is
distinguished
from
the
other
species
by
differences
related
to
OV
(Figs
4,
10C;
Table
5),
from
SAU
by
differences
related
to
CE
(Figs
5,
10D;
Table
5),
and
from
CIR
and
HYP
by
differences
related
to
male
AL
(Fig.
10A;
Table
5).
Pupa.
(No
skin
suitable
for
measuring
avail-
able.)
Last
instar
larva.
Ground
colour
cadmium
orange.
Length
of
flattened
skin
2.1-2.9
mm
(n
=
10).
Mamelons
carrying
anal
papillae
much
small-
er
than
in
most
of
the
other
species
here
discussed
(Fig.
19).
Setal
length,
in
Voc
of
length
of
flattened
skin,
on
outermost
dorsal
papilla
on
prothorax
clearly
shorter
than
in
C1R,
HYP,
PIL
and
REI
(Fig.
20).
Biology.
Host
plant
Inula
salicina.
Of
reared
adults
a
few
emerged
(indoors)
in
the
year
when
respective
larvae
were
collected,
all
others
emerged
not
until
the
following
year.
JA
SCA
ENT.
SCAND.
VOL.
29:3
(1998)
Oligotrophine
gall
midges
in
flower
heads
of
Asteraceae
257
Fig.
13.
Relationships
in
females
of
JAC,
SCA,
etc.,
between
OV,
and
RW,.
Crosses
=
means
and
their
C.i.:s
of
OV,
(vertical
axis)
and
of
RW,
(horizontal
axis).
Rec-
tangles
=
tolerance
intervals
of
OV,
(verti-
cal
axis)
and
of
RW,
(horizontal
axis).
90
85
80
75
HYP
0
70
INU
CIR
SAU
65
60
55
50
'
85 86
87
as
so
90
91
RWc
PIL
UL
92
93
94
95
REI
Material
examined.
From
larvae
collected
in
vii.1977
in
flower
heads
of
lnula
salicina,
3
a,
3
9,
3
pupal
skins,
and
6
skins
of
last
instar
larvae,
SWEDEN:
Oland,
Len-
stad,
E.
Sylven;
from
larva
collected
in
1977,
host
plant
as
above,
1
a,
SWEDEN:
Oland,
the
Great
alvar,
Drostorp,
E.
Sylven;
from
larvae
collected
in
1967,
host
plant
as
above,
I
1
a,
9
9,
and
4
skins
of
last
instar
lar-
vae,
SWEDEN:
Gotland,
Hamra,
E.
Sylven.
Jaapiella
cf.
cirsiicola
Rilbsaamen,
1915
(Code
name:
CIR)
Identification.
Rtibsaamen
(1915)
described
Jaa-
piella
cirsiicola
from
midges
that
he
had
reared
from
larvae
found
in
flower
heads
of
Cirsium
acaule.
Present
examination
of
type
material
of
J.
cirsiicola
(see
Tectotype
designation'
below)
re-
vealed
no
decisive
structural
differences
from
our
specimens
(CIR)
originating
from
flower
heads
of
Cirsium
arvense.
However,
.the
existing
type
mate-
rial
of
J.
cirsiicola
is
by
no
means
in
good
condi-
tion
and
is
not
considered
a
sufficient
basis
for
treating
it
as
identical
with
CIR.
Additional
speci-
mens
reared
from
flower
heads
of
C.
acaule
were
not
available
and
we
therefore
treat
CIR
as
Jaa-
piella
cf.
cirsiicola
Rubsaamen.
Felt
(1911)
described
as
Dasyneura
gibsoni
a
gall
midge
introduced
in
North
America,
probably
from
Europe,
with
larvae
developing
in
flower
heads
of
Cirsium.
Unfortunately
no
material
of
American
origin
of
this
gall
midge
has
been
avail-
able
during
the
present
study.
Maybe
this
gall
midge
is
conspecific
with
the
species
referred
to
here
as
Jaapiella
cf.
cirsiicola.
Jaapiella
narymica
Fedotova,
1996,
infesting
flower
heads
of
Cirsium
serratuloides,
seems
to
be
morphologically
different
from
the
midge
Jaapiel-
la
cf.
cirsiicola
(cp.
Fedotova
1996:
pls
3-4
and
Figs
1,
2
and
5
of
the
present
paper).
Lectotype
designation.
Jaapiella
cirsiicola
Riib-
saamen,
1915.
Type
material:
1
a,
2
9,
1
pupal
skin,
and
2
last
instar
larvae,
Germany:
Triglitz,
from
flower
heads
of
Cirsium
acaule,
leg.
0.
Jaap,
in
Museum
flir
Natur-
kunde,
Berlin.
The
adult
male,
mounted
according
to
the
method
described
by
Sylvan
&
Antipa
Neufeld
(1991),
is
here
designated
as
lectotype.
The
slide
with
this
spec-
imen
has
three
original
labels:
(1)
`Tagesbuch:
IX/257
Coll.
Riibsaamen
Jaapiella
cirsiicola',
(2)
'452',
and
(3)
`ex
coll.
Rilbsaamen'.
A
new
label
has
been
added:
'a
Sy
I
Lectotype
Designated
E.
Sylven
1996'.
Description.
Adult.
No.
of
flagellomeres
12-14
and
11-13
in
males
(n
=9)
and
females
(n
=
17)
re-
258
Sylven,
95
90
85
E.
&
Lindberg,
B.
SCA
JAC
INU
80
IYP
I
75
0
70
CIR
65
El
IL
60
55
50
ENT.
SCAND.
VOL.
29:3
(1998)
Fig.
14.
Relationships
in
females
of
JAC,
SCA,
etc.,
between
OV,
and
RW,..
In
this
case
OK
and
RW„
are
calculated
on
the
basis
of
values
converted
to
correspond
not
to
LW
=
1400
pm
as
in
Fig.
13
but
to
LW
=
1600
pm
(cf.
text).
For
explanation
of
crosses
and
rectangles
see
Fig.13.
SAU
VUL
86
88
90
92
94
96
RWc
spectively.
Male
flagellomeres
III-IV
(Fig.
15)
each
with
neck
about
as
long
as
node
(neck
shorter
than
node
in
the
remaining
species
studied).
Also
distinguished
from
the
other
species
by
differences
related
to
GS
(Figs
3,
10E;
Table
5),
and
from
JAC
by
difference
related
to
OV
(Figs
4,
10C;
Table
5).
Pupa.
Length
of
flattened
skin
1.8-2.2
and
1.6-
2.2
mm
in
males
(n
=
3)
and
females
(n
=
11)
re-
spectively.
Last
instar
larva.
Ground
colour
cadmium
orange.
Length
of
flattened
skin
2.0-3.0
mm
(n
=
13).
Setal
length,
in
%o
of
length
of
flattened
skin,
on
outermost
papilla
on
prothorax
longer
than
in
VUL,
INU
and
SAU
(Fig.
20).
Biology.
Host
plant
Cirsium
arvense.
More
than
one
generation
a
year.
Material
examined.
From
larvae
collected
in
1977
8
cr,
5
9,
and
3
skins
of
last
instar
larvae,
GERMANY:
Schleswig-Holstein,
H.
Meyer;
from
specimens
collect-
ed
in
1968
4
skins
of
last
instar
larvae,
THE
NETHER-
LANDS:
Wageningen
,
W.
Nijveldt;
from
larvae
collect-
ed
in
1959
2
9
and
1
pupal
skin,
SWEDEN:
Scania,
Ven,
E.
Sylven;
from
larvae
collected
in
1987
5
0'
and
7
pupal
skins,
SWEDEN:
Oland,
Northern
peninsula,
E.
Sylven;
from
specimens
collected
in
1967
6
skins
of
last
instar
larvae,
SWEDEN:
Gotland,
Hamra,
E.
Sylven;
from
lar-
vae
collected
in
1959
1
0'
and
2
9,
SWEDEN:
Nordkos-
ter,
Bo,
E.
Sylven;
from
larvae
collected
in
1985
1
0'
and
3
pupal
skins,
SWEDEN:
Sodermanland,
Mariefred,
E.-
Sylven;
from
larvae
collected
in
1978
1
cr
and
6
9,
and
from
larvae
collected
in
1987
1
cr,
4
9
and
4
pupal
skins,
SWEDEN:
Uppland,
Danderyd,
E.
Sylven.
Jaapiella
sp.
on
Centaurea
jacea
and
C.
scabiosa
(Code
names:
JAC,
SCA)
Identification.
Kieffer
(1909)
described
Perrisia
miki
(now
referred
to
Dasineura,
see
Skuhrava
1986)
on
the
basis
of
larvae
from
deformed
flower
heads
of
Centaurea
scabiosa.
As
shown
in
a
draw-
ing
by
Mik
(1896)
the
larval
spatula
of
Kieffer's
species
has
a
pair
of
sharply
pointed
anterior
lobes.
The
midge
with
the
code
name
SCA
also
develops
in
flower
heads
of
Centaurea
scabiosa
but
has
the
anterior
part
of
the
larval
spatula
equipped
with
lobes
that
are
weak
or
even
missing
(Fig.
17).
Fedotova
(1995)
described
Jaapiella
adpressae
and
J.
ruthenicae
from
material
of
gall
midges
in-
festing
flower
heads
of
Centaurea
adpressa
and
C.
ruthenica
respectively.
Unfortunately,
the
original
descriptions
contain
no
information
about
the
ENT.
SCAND.
VOL.
29:3
(1998)
Oligotrophine
gall
midges
in
flower
heads
of
Asteraceae
259
IIr
J
Il
100
pm
8
T
Vf
VII
VIII
200
Am
Fig.
15.
Adult
structures
in
VUL
(left),
CIR
(middle),
and
REI
(right
column).
From
above:
mouth
parts
and
palpus,
cephalic
view
(scale
100
um);
flagellomeres
III-IV
in
male
and
female
respectively,
ventral
view
(scale
100
um);
tar-
sal
tip,
lateral
view
(scale
25
um);
male
uromeres
VI-VIII,
dorsal
view
(scale
200
um).
(e)
empodium;
(h)
hair
sock-
et;
(lb)
labrum;
(11)
labial
lobe;
(pa)
palpifer;
(sc)
scale
sockets.
length
of
OV
or
the
shape
of
the
larval
spatula.
JAC
and
SCA,
reared
from
Centaurea
jacea
and
C.
scabiosa
respectively,
may
well
be
conspecific,
but
this
cannot
be
decided
on
the
basis
of
available
data.
Description.
Adult.
No.
of
flagellomeres
in
JAC
13
and
12-13
in
males
(n
=
2)
and
females
(n
=
14)
respectively,
in
SCA
12-13
and
12-14
in
males
(n
=
6)
and
females
(n
=
7)
respectively.
OV
extraor-
dinarily
long
compared
with
that
in
the
other
spe-
cies
(Figs
4,
10C;
Table
5).
Pupa.
Length
of
flattened
skin
in
JAC
2.2-2.3
mm
in
females
(n
=
2),
and
in
SCA
1.8-2.2
and
2.3-2.7
mm
in
males
(n
=
4)
and
females
(n
=
3)
respectively.
Last
instar
larva.
Ground
colour
cadmium
orange.
Length
of
flattened
skin
in
JAC
2.5-3.1
mm
(n
=
11),
and
in
SCA
2.3-3.5
nun
(n
=15).
Spatula
in
front
with
weak
incision
separating
two
weak
lobes,
or
with
no
incision
at
all
(Fig.
17)
making
its
appearance
markedly
different
from
that
in
the
other
species
(Fig.
17).
For
length
of
se-
tae,
see
Fig.
20.
Variation.
Minor
differences
between
JAC
and
SCA
were
observed,
in
particular
in
values
related
to
RW
(Fig.
9B;
Table
5),
and
in
relative
length
of
certain
setae
on
the
last
instar
larva
(Fig.
20).
These
differences
appear
not
to
be
important
enough
to
treat
JAC
and
SCA
as
different
species.
Biology.
Host
plants
Centaurea
jacea
and
C.
scabiosa.
260
Sylven,
E.
&
Lindberg,
B.
ENT.
SCAND.
VOL.
29:3
(1998)
1
r
ti
id
IOU
ion
Fig.
16.
Pupal
structures
in
VUL
(upper),
HYP
(middle),
and
REt
(lower
row).
From
left
to
right:
right
half
of
front
part,
dorsal
view;
prothoracic
horn,
dorsal
view;
and
detail
of
left
half
of
uromere
VII,
dorsal
view.
(ec)
exterior
ce-
phalic
seta;
(ed)
exterior
dorsal
papilla;
(id)
interior
dorsal
papilla;
(ma)
macrospines;
(mi)
microspines;
(pr)
protu-
berance.
All
to
scale.
Pr
1C
Material
examined.
From
larvae
collected
in
vii.1975
in
flower
heads
of
Centaurea
jacea,
1
cr,
2
9
and
9
skins
of
last
instar
larvae,
SWEDEN,
Oland,
Sjostorp,
E.
Sylven;
from
larvae
collected
in
vii
1989,
on
same
host
plant
and
in
same
locality
as
above,
2
cr,
7
9,
8
pupal
skins,
and
6
skins
of
last
instar
larvae,
E.
Sylven;
from
larvae
collected
in
viii.1978,
host
plant
as
above,
2
skins
of
last
instar
larvae,
SWEDEN:
Oland,
Kalkstad,
E.
Sylven;
from
larvae
collected
in
viii.1977,
host
plant
as
above,
2
9,
SWEDEN:
Uppland,
Bogesundslandet,
Sundby,
E.
Sylven;
from
larvae
collected
in
viii
1977,
host
plant
as
above,
5
9
and
2
pupal
skins,
SWEDEN,
Uppland,
Danderyd,
E.
Sylven.
From
larvae
collected
in
vii.1983
in
flower
heads
of
Centaurea
scabiosa,
3
or,
4
9,
2
pupal
skins,
and
11
skins
of
last
instar
larvae,
SWEDEN:
Oland,
Byxelkrok,
E.
Sylven;
from
larvae
collected
in
vii.1987
or
(1
larva)
in
vii
1989,
host
plant
as
above,
5
cr,
10
9
,
1
pupal
skin,
and
4
skins
of
last
instar
larvae,
SWEDEN:
Oland,
Sjostorp,
E.
Sylven.
Discussion
In
view
of
the
high
morphological
conformity
at
the
species
level
in
various
cecidomyiid
genera
it
is
surprising,
that
only
a
few
morphometric
studies
on
gall
midges
are
found
in
the
literature:
Harris
&
Yukawa
(1980)
succeeded
in
distinguishing
mor-
phometrically
between
two
structurally
similar
gall
midge
species
of
the
genus
Contarinia.
Ros-
kam
&
Kofman
(1987)
and
Roskam
&
Zandee
(1992)
attempted
to
elucidate
the
obviously
very
complicated
paths
in
the
evolution
within
the
ge-
ENT.
SCAND.
VOL.
29:3
(1998)
Oligotrophine
gall
midges
in
flower
heads
of
Asteraceae
261
JAC
SCA
't
VUL
Y2
Fig.
17.
Spatula
sternalis
of
last
instar
larva
in
three
spec-
imens
of
JAC,
SAU,
etc.
Scale
line
corresponding
to
2%
of
length
of
emptied
and
flattened
skin.
CIR
CIR
11
ll
T
REI
HYP
PIL
7
,
SAU
Dasineura,
largely
on
the
basis
of
morphomet-
ric
items.
Sylven
&
Carlbacker
(1981,
1983)
ana-
lysed
certain
allometric
conditions
in
Cecidomyii-
dae,
a
subject
not
previously
discussed
in
the
liter-
ature
for
this
group
of
insects.
In
the
present
work
on
some
gall
midges
of
the
supertribe
Lasiopteridi
that
possesss
few,
if
any,
qualitative
external
differences
between
species,
a
JAC
SCA
morphometric
approach
taking
into
account
intra-
specific
allometric
deviations
has
been
successful-
ly
tested.
A
prerequisite
for
obtaining
reliable
results
in
morphometric
examinations
of
gall
midges
is,
of
course,
that
the
specimens
are
carefully
prepared.
The
uniform
mounting
technique
applied
in
the
present
study
(see
above
under
the
heading
'Mate-
Fig.
18.
Anterior
ventral
pa-
pillae
with
surrounding
areas
in
last
instar
larva
on
right
half
of
uromere
VII,
ventral
view.
(av)
anterior
ventral
papilla;
(ml)
mamelon.
All
to
scale.
HYP
A
-
, t
INU
CIR
PR,
)
REI
VUL
SAU
M.
,
1
00
ign
No.
o
I
specimens
TIP'
Cr=r1L'ITI
M
TO
1
15
-0
ors
UVIIPV
,4
UVIU
D
._242=E1r,
16.1
ENT.
SCAND.
VOL.
29:3
(1998)
Oligotrophine
gall
midges
in
flower
heads
of
Asteraceae
263
VUL
IN
U
SA
U
JAC
SCA
CIR
HYP
PIL
R
El
VUL
/NU
SA
U
JAC
SCA
CIR
HYP
Pt1
El
TI
OD
10
15
1
5
10
15
1
5
10
15
5
10
15
1
5
10
15
1
5
10
15
Fig.
20.
Last
instar
larva.
Setal
length
in
7
-
of
length
of
emptied
and
flattened
skin.
White
bars
and
black
bars
=
setae
in
specimens
with
skin
length
<
2.6
mm
and
>
2.6
mm
respectively.
Bars
turned
upwards
and
bars
turned
downwards
=
setae
on
left
and
right
half
of
body
respectively.
T
I
OD
=
seta
on
left
or
right
outermost
dorsal
papilla
on
prothor-
ax;
T
III
OD
and
U
VII
OD
=
ditto
on
metathorax
and
uromere
VII
respectively;
T
III
IP
=
ditto
on
inner
pleural
pa-
pilla
on
metathorax;
U
VII
PV
=
ditto
on
posterior
ventral
papilla
on
uromere
VII;
U
VIII
D
=
ditto
on
dorsal
papilla
on
uromere
VIII.
allometric
deviations
in
morphometric
items
should
not
to
be
ignored.
Even
if
no
significant
de-
viation
from
proportionality
could
be
ascertained
in
many
of
the
analyses
presented
in
Figs
9-10,
still
more
than
50%
of
the
comparisons
between
two
species
based
on
corrected
values
(see
Table
264
Sylven,
E.
&
Lindberg,
B.
ENT.
SCAND.
VOL.
29:3
(1998)
5)
are
influenced
by
significant
deviation
from
isometry
in
one
or
both
species.
Janzon
(1986),
in
a
morphometric
study
on
chalcidoid
Hymenop-
tera,
adopted
the
methods
used
by
Sylven
&
Carlbacker
(1981,
1983).
Several
of
his
analyses
revealed
examples
of
negative
allometry.
It
would
be
of
interest
to
perform
morphometric
studies,
including
analyses
of
allometric
condi-
tions,
in
other
groups
of
Cecidomyiidae
such
as
Lestremiinae
and
Porricondylinae.
In
these
sub-
families
species
identification
is
often
only
pos-
sible
in
the
male
sex.
The
morphometric
approach
may
be
of
help
in
distinguishing
females
of
many
species.
The
present
study
indicates
a
high
host
specific-
ity
for
all
of
the
recognized
species
of
the
gall
midges.
This
is
in
accordance
with
the
results
of
various
other
studies
dealing
with
host
ranges
of
phytophagous
gall
midges
(e.g.,
Barnes
1953,
Roskam
1977,
SyIv&
1979,
Gagne
&
Marohasy
1993,
Sylven
&
Lovgren
1995,
Ahman
&
Lovgren
1995).
It
is
true
that
the
identity
of
phytophagous
gall
midges
can
often
be
determined
on
the
basis
of
their
host
plant
and
the
kind
of
gall
they
induce.
However,
caution
is
recommended
based
on
the
fact
that
many
phytophagous
gall
midges
develop
as
inquilines
in
galls
produced
by
other
gall
midg-
es.
Such
associations
occur
even
between
closely
related
gall
midges,
e.g.,
between
certain
oligotro-
phine
species
infesting
Salix
(see
Stelter
1982).
For
identification
of
phytophagous
gall
midges
causing
no
obvious
galls,
as
in
the
species
dealt
with
in
the
present
study,
morphological
examina-
tion
of
the
midges
themselves
is
particularly
im-
portant.
Acknowledgments
Professor
Bengt
Jonsell
(Stockholm)
and
Professor
Ber-
til
Nordenstam
(Stockholm)
assisted
in
host
plant
iden-
tification
(see
Table
I).
Mrs
Ellen
Nyholm
(Taby,
Swe-
den)
and
Mrs
Maria
Antipa
Neufeld
(Stockholm)
pro-
duced
microscopic
slides
used
in
the
study,
Nyholm
also
most
of
the
drawings.
In
addition,
Mrs
Gunnel
Seller-
holm
(Stockholm)
contributed
technical
assistance.
Dr
Hans
Meyer
(Bordesholm,
Germany)
and
Dr
Wim
Nij-
veldt
(Wageningen)
placed
at
our
disposal
certain
speci-
mens
included
in
the
material
analysed.
Dr
Anders
Big-
nert
(Stockholm)
and
Professor
Bengt
Enflo
(Stock-
holm)
kindly
commented
on
the
statistical
methods
used.
Dr
Keith
Harris
(Woking,
U.K.)
kindly
read
and
critically
examined
the
manuscript.
All
the
above
per-
sons
are
gratefully
acknowledged.
References
Barnes,
H.
F.
1946.
Gal]
midges
of
economic
impor-
tance.
Vol.
I.
Gall
midges
of
root
and
vegetable
crops.
104
pp.
London.
-
1953.
The
biological
approach
to
the
species
problem
in
gall
midges
(Dipt.,
Cecidomyidae).
Annales
entom-
ologici
fennici
19:
1-24.
Fedotova,
Z.
A.
1993.
New
species
of
flower
gall
midges
(Diptera,
Cecidomyiidae)
from
the
mountains
of
Ka-
zakhstan.
Zoologicheskii
Zhurnal
72:
76-90.
-
1995.
New
gall-midge
species
(Diptera,
Cecidomyii-
dae)
from
Asteraceae
host
plants
of
Kazakhstan.
Vest-
nik
Zoologii
5-6:
18-24.
-
1996.
New
species
of
gall-midges
(Diptera,
Cecido-
myiidae)
from
flower
heads
of
Asteraceae
in
Kazakh-
stan.
Entomologicheskoe
Obozrenie
75:
192-207.
Felt,
E.
R
1911.
New
species
of
gall
midges.
Journal
of
Economic
Entomology
4:
476-484.
Gagne,
R.
J.
1994.
The
gall
midges
of
the
Neotropical
Region.
xiv
+
352
pp.
Ithaca.
Gagne,
R.
J.,
Harris,
K.
M.,
Skuhrava,
M.,
Solinas,
M.
&
Sylven,
E.
1997.
Dasineura
Rondani,
1840
(Insecta,
Diptera):
proposed
designation
of
Tipula
sisymbrii
Schrank,
1803
as
the
type
species.
Bulletin
of
Zoolog-
ical
Nomenclature
54:
92-94.
Gagne,
R.
J.
&
Marohasy,
J.
1993.
The
gall
midges
(Dip-
tera:
Cecidomyiidae)
of
Acacia
spp.
(Mimosaceae)
in
Kenya.
Insecta
mundi
7:
77
1
24.
Harris,
K.
M.
&
Yukawa,
J.
1980.
The
taxonomic
status
of
the
Japanese
citrus
flower-bud
midge,
Contarinia
okadai
(Miyoshi)
comb.
n.,
and
of
the
citrus
blossom
midge,
C.
citri
Barnes
(Diptera:
Cecidomyiidae).
Bul-
letin
Entomological
Research
70:
285-287.
Janzon,
L.-A.
1986.
Morphometric
studies
of
some
Pter-
omalus
species
(Hymenoptera:
Chalcidoidea)
with
emphasis
on
allometric
relationships,
or:
Are
ratios
re-
liable
in
chalcid
taxonomy?
Systematic
Entomology
11:75-82.
Kicffcr,
J. J.
1888.
Beitrage
zur
Kenntniss
der
Gallmiicken.
Entomologische
Nachrichten
2
310-314.
-
1909.
Contributions
a
la
connaissance
des
insectes
gal-
licoles.
Bulletin
de
la
Societe
d'Histoire
Naturelle
de
Metz
28:
1-35.
Mik,
J.
1896.
Eine
neue
Cecidomyiden-Galle
auf
Cen-
taurea
Scabiosa
L.
Wiener
Entomologische
Zeitung
15:
292-294.
MOM,
E.
1955.
Beitrage
zur
Systematik
der
Larven
dcr
ltonididae
(=
Cecidomyiidae,
Diptera).
1.
Teil.
Porri-
condylinae
und
Itonidinae
Mittelcuropas.
Zoologica,
Stuttgart
38
(105):
1-247.
Rohlf,
F.
J.
&
Sokal,
R.
R.
1969.
Statistical
tables.
xi
+
253
pp.
San
Francisco.
Roskam,
J.
C.
1985.
Evolutionary
patterns
in
gall
midge
-host
plant
associations
(Diptera,
Cecidomyiidae).
TUdschrift
voor
Entomologie
128:
193-213.
Roskam,
H.
C.
&
Kofman
S.
1987.
Phylogenetic
studies
in
Dasineurini
(Diptera:
Cecidomyiidae).
Cecidologia
Internationale
8:
55-70.
Rfibsaamen,
E.
H.
1915.
Cecidomyidenstudien.
IV.
Re-
vision
der
deutschen
Oligotropharien
und
Lasiopter-
arien
nebst
Beschreibung
neuer
Arten.
Sitzungsbe-
richte
Gesellschaft
Naturforschender
Freunde
zu
Ber-
lin
10:
485-567.
Skuhrava,
M.
1986.
Cecidomyiidae,
pp.
72-297
in
Soos
ENT.
SCAND.
VOL.
29:3
(1998)
Oligotrophine
gall
midges
in'
flower
heads
of
Asteraceae
265
&
Papp
(Eds):
Catalogue
of
Palaearctic
Diptera.
Vol.
4.
Sciaridae
Anisopodidae.
Amsterdam
&
Budapest.
Sokal,
R.
R.
&
Rohlf,
F.
J.
1969.
Biometry.
The
princi-
ples
and
practise
of
statistics
in
biological
research.
xxi
+
776
pp.
11995,
3rd
ed.,
xix
+
887
pp.]
San
Fran-
cisco.
Stelter
H.
1982.
Untersuchungen
Ober
Gallmucken
XXV.
Zur
Differenzierung
von
Rhabdophaga
hetero-
bia
(H.
Loew,
1850),
Rh.
terminalis
(H.
Loew,
1850)
and
verwandter
Arten.
Entomologische
Abhandlun-
gen
Staatliches
Museum
fur
Tierkunde
in
Dresden
46:
69-84.
Sylven,
E.
1975.
Study
on
relationships
between
habits
and
external
structures
in
Oligotrophidi
larvae
(Dip-
tera,
Cecidomyiidae).
Zoologica
scripta
4:
55-92.
1979.
Gall
midges
(Diptera,
Ceeidomyiidae)
as
plant
taxonomists.
Symbolae
botanicae
upsalienses
22:
62-
69.
Sylven,
E.
&
Antipa
Neufeld,
M.
1991.
A
preparation
method
for
gall
midges.
Entomologisk
Tidskrift
112:
153-155.
Sylven,
E.
&
Carlbacker,
U.
1981.
Morphometric
studies
on
Oligotrophini
adults
(Diptera:
Cecidomyiidae)
in-
cluding
an
attempt
to
correct
for
allometric
devia-
tions.
Entomologica
scandinavica,
Supplement
15:
185-210.
1983.
Allometric
relationships
concerning
antenna
and
wing
in
Oligotrophini
(Diptera:
Cecidomyiidae).
Acta
entomologica
fennica
42:
90-102.
Sylven,
E.
&
Lovgren,
L.
1995.
Dasineura
ingeris
sp.
n.
(Diptera:
Cecidomyiidae)
on
Salix
viminalis
in
Swe-
den,
including
comparisons
with
some
other
Dasineu-
ra
species
on
Salix.
Systematic.
Entomology
20:
59-71.
Sylven,
E.
&
Tas0s-Duque,
R.
1993.
Adaptive,
taxo-
nomic,
and
phylogenetic
aspects
of
female
abdominal
features
in
Oligotrophini
(Diptera:
Cecidomyiidae),
and
four
new
Dasineura
species
from
the
Western
Palearctic.
Zoologica
scripta
22:
277-298.
Ahman,
I.
&
Lovgren
L.
1995.
Host
relationships
of
Dasineura
ingeris:
oviposition
preferences
and
juve-
nile
survival
on
clones
of
Salix
viminalis,
other
Salix
species
and
hybrids.
Etztomologia
experimentalis
et
applicata
77:
23-30.
Revised
manuscript accepted
May
1998.