Oenothera speciosa versus Macroglossum stellatarum: killing beauty

Abstract

Hovering and dead individuals of the diurnal hawk-moth Macroglossum stellatarum (Linnaeus, 1758) (Lepidoptera: Sphingidae) were found with proboscides got stuck into flowers of the ornamental plant Oenothera speciosa Nutt (Onagraceae). The phenomenon was observed in several locations in Bulgaria where the plant has been introduced. Microscopic examination revealed that the reason for this unusual interaction is pubescence of thick-walled basiscopically oriented trichomes in the basal part of the hypanthium and style of the plant. When a foraging moth inserts its proboscis into this area, the tips of the trichomes are inserted into the transverse grooves of proboscis and hamper its back movement. As a result the moths are suspended for a long time, sometimes until death. Other trapped moth species were also observed but they always effected self-release. This plant–insect interaction is also a conservation issue as an estimation of its impact on wild insect populations is lacking.

Full text

UNCORRECTED PROOF
Journal : Large 11829 Article No : 9588 Pages : 6MS Code : APIS-D-17-00142 Dispatch : 5-12-2017
Vol.:(0123456789)
1 3
Arthropod-Plant Interactions
https://doi.org/10.1007/s11829-017-9588-3
ORIGINAL PAPER
Oenothera speciosa versusMacroglossum stellatarum: killing beauty
BoyanZlatkov1,2· StoyanBeshkov2· TsvetaGaneva3
Received: 9 October 2017 / Accepted: 2 December 2017
© Springer Science+Business Media B.V., part of Springer Nature 2017
Abstract
Hovering and dead individuals of the diurnal hawk-moth Macroglossum stellatarum (Linnaeus, 1758) (Lepidoptera: Sphin-
gidae) were found with proboscides got stuck into flowers of the ornamental plant Oenothera speciosa Nutt (Onagraceae).
The phenomenon was observed in several locations in Bulgaria where the plant has been introduced. Microscopic examina-
tion revealed that the reason for this unusual interaction is pubescence of thick-walled basiscopically oriented trichomes in
the basal part of the hypanthium and style of the plant. When a foraging moth inserts its proboscis into this area, the tips of
the trichomes are inserted into the transverse grooves of proboscis and hamper its back movement. As a result the moths are
suspended for a long time, sometimes until death. Other trapped moth species were also observed but they always effected
self-release. This plant–insect interaction is also a conservation issue as an estimation of its impact on wild insect popula-
tions is lacking.
Keywords Introduced plant· Trapping trichomal zone· Stuck proboscis· Trapped hawk-moths
Introduction
The genus Oenothera comprises 120–200 species (Gregory
1963; Dietrich etal. 1997; Mabberley 1997). Many of them
are pollinated by large nocturnal moths, especially Sphin-
gidae, and have typical adaptations, i.e. they are sphingo-
philous: deep hypanthium, pale corolla colour, spreading
four-lobed stigmas elevated above the anthers, fragrance and
nectar (Gregory 1963, 1964; Raven 1979; Raju etal. 2004).
All these features are present in O. speciosa. The plant orig-
inates from the southern part of North America and was
introduced as an ornamental plant in Europe, including Bul-
garia. Its flowers are open throughout most of the daytime
and also at night, except for the hottest hours and in its native
range it is visited by various insects (Wolin etal. 1984).
Many hawk-moths (Lepidoptera: Sphingidae) are
renowned for their long proboscides which are adapted for
feeding from specific flowering plants; some are specialised
to the level of feeding from, and hence pollinating, only a
single plant species. However, most sphingids are general-
ists and visit vast range of plants. A typical representative of
the second group is Macroglossum stellatarum (Linnaeus,
1758) (Sphingidae), native to Europe, and well-known diur-
nal visitor of many plant species. An interesting interaction
has been observed in Bulgaria: When this moth visits flow-
ers of the introduced ornamental plant Oenothera speciosa
Nutt. (Onagraceae), its proboscis becomes physically stuck
into the flower and the moth cannot escape, usually until
death. Similar observations in France are published in the
social networks and some journals (e.g. Benéton 2009).
Other sphingids and noctuids were also observed visiting,
and some of them temporally trapped by, the flowers of the
same plant. We could not find any scientific explanation of
the phenomenon, therefore the aim of the study is to find a
reason why the proboscis of M. stellatarum gets stuck inside
the flower of O. speciosa.
*Boyan Zlatkov
bzlatkov@gmail.com
1 Institute ofBiodiversity andEcosystem Research, Bulgarian
Academy ofSciences, 1 Tsar Osvoboditel Blvd., 1000Sofia,
Bulgaria
2 National Museum ofNatural History, Bulgarian Academy
ofSciences, 1 Tsar Osvoboditel Blvd., 1000Sofia, Bulgaria
3 Faculty ofBiology, Sofia University St. Kliment Ohridski, 8
Dragan Tsankov Blvd., 1164Sofia, Bulgaria
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
A1
A2
A3
A4
A5
A6
A7
A8
A9
This is an accepted manuscript. The final publication is
available at https://link.springer.com/article/10.1007/s11829-017-9588-3

UNCORRECTED PROOF
Journal : Large 11829 Article No : 9588 Pages : 6MS Code : APIS-D-17-00142 Dispatch : 5-12-2017
B.Zlatkov et al.
1 3
Materials andmethods
The main observation on the moths and collecting of mate-
rial was performed in September 2016 and 2017 in a pri-
vate yard in Milanovo, in the Sofia region of Bulgaria, at
an altitude of 860m. Additional data came from a private
garden in Knezha in the Pleven region at 120m altitude,
and Kenana city park in Haskovo, Bulgaria, at 250m alti-
tude. Living and dead moths were collected together with
flowers. Five living and five dead M. stellatarum with pro-
boscides stuck into flowers were preserved in 70% etha-
nol. In the laboratory, the proboscides were cut off near
the palps. From the plants, the petals, sepals and part of
the ovary were removed and the remaining part of hyp-
anthium and style, containing the amputated proboscis,
were processed following a standard histological proce-
dure: dehydration with ethanol, clearing with xylene and
embedding in wax media Paraplast Plus® (Sigma-Aldrich).
Five blocks were prepared, two of them were sectioned
transversely and three longitudinally with a microtome at
10µm thickness, then deparaffinised, stained with 0.1%
toluidine blue aqueous solution and mounted with Canada
balsam on glass slides. The lengths of proboscides from
the palps to the tip were measured on ethanol-preserved
specimens of five M. stellatarum and on dry specimens,
rehydrated with hot water or 10% KOH solution, of
Autographa gamma (Linnaeus, 1758), Helicoverpa armig-
era (Hübner, [1808]) (Noctuidae), Agrius convolvuli (Lin-
naeus, 1758), Hyles livornica (Esper, 1780) and Sphinx
pinastri Linnaeus, 1758 (Sphingidae)—one specimen of
each. The lengths of proboscides of the sphingids were
measured with a ruler. All other measurements in the study
were performed with microscope ocular scales. The data
from the measurements were statistically processed and
the means ± standard deviation are present in the text and
Table1. Most of the trapped individuals were released
by the observers without determination of their sex. The
histological sections were observed and photographed
on an Amplival (Carl Zeiss Jena) compound microscope
equipped with apochromatic objectives, photo eyepiece
MF Messprojektiv K 4:1 and Canon Eos 70D digital cam-
era. Additionally, flowers with stuck proboscides were
dissected and photographed insitu under a Stemi 2000-C
stereomicroscope with an AxioCam ERc5s (Carl Zeiss)
digital camera. Macrophotographs were taken with a Sony
DSChX400v digital camera. The photographs were pro-
cessed with Digital Photo Professional 4 (Canon) software.
Results
Hovering (Fig.1a), exhausted and dead individuals of M.
stellatarum suspending on flowers of O. speciosa (Fig.1b)
were observed. Close examination showed that the moths
had their proboscides got stuck into the flowers. Apparently,
the captured moths were struggling to get loose from the
flowers, which eventually led to exhaustion and sometimes
death. On several occasions the moths self-released in a cou-
ple of minutes when picked-up together with flowers.
Examination under a stereomicroscope showed that the
proboscides had been inserted with their tips reaching the
bottom of the hypanthium (Fig.1c). A dense cover of tri-
chomes on the style and hypanthium were suspected of
hampering a reflex movement of the proboscis (Fig.1d).
Microtome sections clearly supported this assumption. The
hypanthium length of O. speciosa is 16–21mm (mean
18.80 ± 1.92, n = 5). It is subcylindrical with the narrow-
est area at the base. The hypanthium has four symmetri-
cally arranged longitudinal grooves in which the proboscis
can be inserted (Fig.2a). The internal basal part of the
hypanthium and the corresponding part of the style are
densely pubescent (Figs.1d, 2e). The base of the tube
is a glabrous zone with length of 1.2–1.6mm (mean
1.36 ± 0.15, n = 5) where the nectar is discharged. The tri-
chomes are non-glandular unicellular, basiscopic (oriented
toward the base), with a thick wall and slightly sigmoid
form. The length of the hypanthium trichomes (Fig.2c)
varies between 110 and 135μm (mean 123.00 ± 7.97,
n = 15). The trichomes of the style (Fig.2d) are shorter,
100–120μm (mean 110.3 ± 6.67, n = 15). The pubescent
zone is with length of ca. 3.9–4.3mm; (mean 4.12 ± 0.18,
n = 5). The length of the proboscis of M. stellatarum is
24–27 mm (mean 25.40 ± 1.34, n = 5). The diameters
(sagittal and frontal) of the proboscis in the middle of
the contact area with trichomes are ca. 156μm (mean
Table 1 Proboscis length and
interaction of the observed moth
species visiting O. speciosa
Species Proboscis length (mm) Interaction
Sphinx pinastri 31 (n = 1) Trapped, self-released
Agrius convolvuli 94 (n = 1) Trapped, self-released
Macroglossum stellatarum 25.40 ± 1.34 (24–27) (n = 5) Trapped, dead or self-released
Hyles livornica 29 (n = 1) Trapped, self-released
Autographa gamma 16 (n = 1) Not trapped
Helicoverpa armigera 10 (n = 1) Not trapped
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132

UNCORRECTED PROOF
Journal : Large 11829 Article No : 9588 Pages : 6MS Code : APIS-D-17-00142 Dispatch : 5-12-2017
Oenothera speciosa versusMacroglossum stellatarum: killing beauty
1 3
155.6 ± 5.11, n = 15) and 349 μm (mean 348.5 ± 4.74,
n = 10), respectively. The distance between the tips of the
hypanthium and style trichomes varies between 50 and
100μm (mean 81.2 ± 23.93, n = 5). It is clear that the fron-
tal diameter of the proboscis is larger than the mean dis-
tance between the opposing trichomes in the tube grooves,
therefore the proboscis can interact with trichomes from
at least two sides. The proboscis has numerous transverse
grooves separating cuticular annulations, which fit well
the trichome tips (Fig.2c). When a moth inserts its pro-
boscis into the flower, the tips of the trichomes easily get
into the grooves, but reflex movement of the proboscis is
hampered by the basiscopic orientation of the trichomes.
In this way, the proboscis gets stuck into the hypanthium
and the moth is unable to break loose. Occasionally, some
moths can self-release without external assistance. Exami-
nation of the hypanthia failed to find the remaining part
of the proboscis of the self-released moths, indicating that
they survived the event intact. The ratio of captured/self-
released moths was not estimated, because the observers
in most cases released them forcibly. Differences between
sexes in visiting, trapping and self-release from the flowers
were not detected because the sex of the individuals was
not determined during observation.
Nocturnal hawk-moth species stuck to flowers were also
observed, but they always escaped without external assis-
tance: A. convolvuli, H. livornica, S. pinastri, all of them
considerably larger than M. stellatarum and with longer pro-
boscides (Table1). For smaller moths, such as A. gamma
and H. armigera and diurnal Hymenoptera as Xylocopa vio-
lacea (Linnaeus, 1758) and Apis mellifera Linnaeus, 1758
(Apidae) visiting the flowers appeared trouble-free.
Discussion
Pollination by insects is important for O. speciosa, as it is
partially self-incompatible (Wolin etal. 1984). The anthers
and stigma of the flower are raised at long distance from
the petals, in this way feeding moths can touch them. This
effect can be intensified if a moth hovers for a prolonged
time around the flower. The structure of trichomes may
be related to different pollinating strategies as different
Oenothera species have evolved different trichomes: long
Fig. 1 Macroglossum stellatarum with proboscides got stuck into
flowers of Oenothera speciosa. a Hovering moth. b Dead moth. c
Longitudinal section of flower with proboscis completely inserted. d
A detail of trichomal zone with proboscis. 1 proboscis, 2 style 3 hyp-
anthium, 4 trichomal zone, 5 nectaries. Scales b 10mm, c 5mm, d
0.5mm
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173

UNCORRECTED PROOF
Journal : Large 11829 Article No : 9588 Pages : 6MS Code : APIS-D-17-00142 Dispatch : 5-12-2017
B.Zlatkov et al.
1 3
and lanate; short, straight and perpendicular to the longi-
tudinal axis; short and basiscopic (Gregory 1963, 1964).
However, putatively one can conclude that the flowers of
O. speciosa have co-evolved adaptations in the form of
trapping trichomes to ensure pollination by means of tem-
porary suspension of the pollinator.
The length of the proboscis of M. stellatarum is greater
than the length of the hypanthium of O. speciosa, which
means that the moth can reach the trichomal zone and nec-
taries. The diameters of the proboscis are larger than the
mean distance between the opposing trichomes in the hypan-
thium. When the tips of the trichomes enter the annulations
Fig. 2 Histological sections of the trichomal zone of hypanthium
with proboscis inserted. a Cross section with proboscis in a groove
of hypanthium. b Longitudinal section of hypanthium with proboscis.
c A detail of b; note that the tip of hypanthium trichome fits well an
annulation of proboscis. d Style trichome. e Longitudinal section of
a groove of the trapping trichomal zone, left side style surface, right
side hypanthium internal surface. Scales a 250μm, b, e 100μm, c, d
20μm
174
175
176
177
178
179
180
181
182
183
184
185

UNCORRECTED PROOF
Journal : Large 11829 Article No : 9588 Pages : 6MS Code : APIS-D-17-00142 Dispatch : 5-12-2017
Oenothera speciosa versusMacroglossum stellatarum: killing beauty
1 3
of the proboscis, they act as anchors and back movement
of the proboscis needs considerable effort, which is usually
unattainable for M. stellatarum. Though a superior flier, M.
stellatarum is not powerful enough to self-release which
leads to exhaustion and eventually death of the moth. Larger
and more powerful hawk-moths are also trapped, probably
by the same mechanism; however, they can overcome the
“flower trap” (Table1). Moths with shorter and more slender
proboscides such as noctuids are not captured at all. Per-
haps the diameter of their proboscides is not large enough
to interact effectively with the trichomes. One of the species
observed can reach the trichomal area and nectar of at least
some flowers (the hypanthium in part of the flowers is longer
than the proboscis of A. gamma). The proboscis of the other
noctuid (H. armigera) is too short, it cannot reach the nectar
and one can speculate that it is an accidental visitor attracted
by the scent. The observed Apidae also were not trapped by
the flowers; perhaps they use only pollen from O. speciosa.
They have shorter proboscis (in A. mellifera the functional
length is 7mm, Waddington and Herbst 1987) with different
anatomy, therefore they can hardly be trapped. Shortly after
picking-up a flower together with a living moth, the probos-
cis is released probably because of loss of turgor pressure.
The presence of adaptations for temporary suspending the
moths can be an evolutionary advantage, as it can affect the
pollination positively. Perhaps this could be the case with
large and powerful hawk-moths in the native range of the
plant but not with M. stellatarum. Moreover, it is not clear
whether a moth trapped and self-released from a flower will
visit the same plant species again, as the hawk-moths clearly
demonstrate learning by experience (Raguso and Willis
2003). Questions relating to the pollination of O. speciosa
outside its native range deserve further investigation.
Though several moth species were captured by the flow-
ers of O. speciosa, the prevailing one is M. stellatarum. The
specifics in the anatomy of the moth are the obvious reason,
but other factors can also contribute: M. stellatarum is a
common and apparently numerous species in Bulgaria forag-
ing at daytime, when the flowers of O. speciosa are open. It
is interesting to note that in its native range, O. speciosa is
regularly visited by Lepidoptera, all of them Papilionoidea
and not Sphingidae (Wolin etal. 1984). These authors
claimed that the main pollinator is Apis mellifera. Most
Papilionoidea were ineffective pollinators as they did not
touch the stigma, with the exception of one species, Battus
philenor (Linnaeus, 1771), which was considered an impor-
tant cross-pollinator. O. speciosa appears to be rarely visited
(and probably rarely pollinated) by hawk-moths in North
America and no cases of dead moths with trapped probos-
cides are reported. Recently, M. stellatarum was registered
in California, USA (Pittaway 2017). It would be interesting
to search for possible interactions between the moth and O.
speciosa in its native range.
Most of the moths trapped by flowers were observed in
the second half of September of 2016 and 2017, when the
flowering period of the autochthonous vegetation in Bul-
garia was over. It should be underlined that the flowering
period of O. speciosa in the area of observation is consid-
erably extended, from June to the second half of October.
The killing of M. stellatarum by O. speciosa presents
a conservation issue. In our observation, the plant is very
popular and is grown mainly in private gardens throughout
Bulgaria. The phenomenon of moth capturing is presum-
ably common, as it attracted the attention of independent
observers without special knowledge from different areas
of the country. It is impossible to estimate with any degree
of accuracy how many moths were killed by O. speciosa,
but during three days a total of twelve captured moths were
found from a small area with around 50 single flowers.
Therefore we consider that the growing of O. speciosa in
gardens in Bulgaria is undesirable, particularly as there is
no adequate assessment of its true impact on wild popula-
tions of M. stellatarum and other insects. The trapping
of M. stellatarum can be beneficial for other organisms,
however. Benéton (2009) reported Mantis religiosa (Lin-
naeus, 1758) feeding on trapped moths; likely, other preda-
tors may take an advantage of this unusual plant–insect
interaction.
The flowers of O. speciosa are not the only ones trap-
ping moths in nature. Araujia sericifera Brot. (Apocyn-
aceae) is a well-known plant in which flowers various spe-
cies of moths, butterflies and other insects as well meet
their death. The trapping mechanism of the flower is differ-
ent however: the sticky pollinaria attached to the proboscis
wedge it between the anther wings. In this way an insect
is permanently trapped to the flower. In both cases of A.
sericifera and O. speciosa, the killing of the flower visitors
should be treated as a side effect of a specific pollination
strategy. It is interesting to note that A. sericifera was also
introduced to Europe from South America and has become
invasive in several southern European countries (Coombs
and Peter 2010). Regardless the different mechanisms of
trapping, the killing of indigenous insect species by intro-
duced plants during flower visiting certainly deserves fur-
ther assessment of its ecological impact.
Acknowledgements We are most grateful to Vladimir Beshkov (Sofia,
Bulgaria) who pointed our attention to the phenomenon and provided
valuable observations and material. Elena Nankova (Haskovo, Bul-
garia) shared her pictures and data. Teodora Toshova (Sofia, Bulgaria)
also provided important observations. Thanks go also to Ulf Eitsch-
berger (Marktleuthen, Germany) for consultation and providing lit-
erature and Colin W. Plant (Bishops Stortford, Hertfordshire, UK) for
linguistic correction of and comments on the manuscript. The review-
ers made valuable comments which improved the manuscript.
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289

UNCORRECTED PROOF
Journal : Large 11829 Article No : 9588 Pages : 6MS Code : APIS-D-17-00142 Dispatch : 5-12-2017
B.Zlatkov et al.
1 3
References
Benéton M (2009) Un piège mortel pour le moro-sphinx. L’Érable
2009(1):18–20
Coombs G, Peter CI (2010) The invasive ‘mothcatcher’ (Araujia seric-
ifera Brot.; Asclepiadoideae) co-opts native honeybees as its pri-
mary pollinator in South Africa. AoB Plants 2010:plq021
Dietrich W, Wagner WL, Raven PH (1997) Systematics of Oenothera
section Oenothera subsection Oenothera (Onagraceae). Syst Bot
Monogr 50:1–234
Gregory DP (1963) Hawkmoth pollination in the genus Oenothera.
Aliso 5(3):357–384
Gregory DP (1964) Hawkmoth pollination in the genus Oenothera.
Aliso 5(4):385–419
Mabberley DJ (1997) The plant book. Cambridge University Press,
Cambridge
Pittaway AR (2017) Sphingidae of the Western Palaearctic (includ-
ing Europe, North Africa, the Middle East, western Siberia and
western Central Asia). http://tpittaway.tripod.com/sphinx/list.htm.
Accessed 6 Oct 2017
Raguso RA, Willis MA (2003) Hawkmoth pollination in Arizona’s
Sonoran desert: behavioral responses to floral traits. In: Boggs
CL, Watt WB, Ehrlich PR (eds) Butterflies: ecology and evolution
taking flight. Chicago University Press, Chicago
Raju AJS, Rao SP, Ezradanam V, Zafar R, Kalpana PR, Kumari
PK (2004) The hawkmoth Macroglossum gyrans and its inter-
action with some plant species at Visakhapatnam. Zoos’ Print
19(9):1595–1598
Raven PH (1979) A survey of reproductive biology in Onagraceae. NZ
J Bot 17(4):575–593
Waddington KD, Herbst LH (1987) Body size and the functional length
of the proboscis of honey bees. Florida Ent 70(1):124–128
Wolin CL, Galen C, Watkins L (1984) The breeding system and aspects
of pollination effectiveness in Oenothera speciosa (Onagraceae).
Southwest Nat 29(1):15–20
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323