Female oviposition preference
Bruchid females deposited significantly decreasing number of eggs on seeds in the order of H, ANH and NH species (Fig. 1, Kruskal-Wallis ANOVA: H2,140 =48.644, p < 0.001). Females laid 36.9 ± 1.8 (mean ± SE) eggs on H species (N = 22 SCAs), 18.9 ± 1.3 eggs on ANH species (N = 55 SCAs), and 13.4 ± 0.89 eggs on NHs (N = 63 SCAs). The females laid significantly more eggs on larger seeds (Fig. 1, Kruskal-Wallis test: H2,132 =73.057, p < 0.001). Mean mass of H seeds was 340.8 ± 24.0 mg, that of ANHs 235.3 ± 22.9 mg, and that of NHs 69.5 ± 9.7 mg. All 66 plant species (62 leguminous and 4 outgroup species) received eggs, but females laid less than 15 eggs on 55%, less than 30 eggs on 35% and between 30–45 eggs on 10% of plant species. Figure 2 shows the distribution of eggs among leguminous tribes. Not surprisingly, A. obtectus females laid the highest number of eggs on members of the tribe Phaseoleae, where the main hosts are also found. Within Phaseoleae soybeans (Glycine) was the least preferred. Comparable responses were noted to some species within Caesalpinieae, Genisteae, Robinieae, Cicereae and Fabeae (ESM Table 1). Although none of them being a host for the bean weevil, within the tribe of Caesalpinieae Gleditsia delavayi, within Genisteae Laburnum alpinum and L. anagyroides, and within Robinieae Robinia viscosa received high numbers of eggs. Cicer arietinum and Vicia faba are known as occasional hosts. The number of eggs laid on them fell into the medium and high categories, respectively. Among the legumes the lowest numbers of eggs were laid on Vicia tenuifolia, Robinia pseudoacacia and Amorpha fruticosa, but on some members of the Fabeae tribe, comprising Vicia and Lathyrus species, the mean number of eggs was less than 10. Females accepted the pilules incorporated with seed coat as an oviposition substrate and laid comparable number of eggs to control (Table 3).
Table 1
Mean seed mass and seed coat thickness of leguminous plants from which bean weevil (A. obtectus) adults emerged. Glycine max comprises 17, Vigna unguiculata 2, Phaseolus vulgaris 21, Vicia faba 6, Lens culinaris 2, Pisum sativum 27 cultivars, and Lathyrus tuberosus 2 samples
Plant tribe and species1
|
Seed mass2 (mg)
±SE (N)
|
Seed coat thickness3 (mm) ± SE (N)
|
Genisteae
|
|
|
Lupius albus L.
|
197.2 ± 4.1 (46)
|
0.20 ± 0.007 (25)
|
Phaseoleae
|
|
|
Glycine max (L.) Merr.4
|
167.9 ± 0.7 (910)
|
0.09 ± 0.001 (265)
|
Lablab purpureus (L.) Sweet
|
142.2 ± 3.5 (298)
|
0.15 ± 0.004 (24)
|
Vigna unguiculata (L.) Walp.
|
109.6 ± 5.2 (70)
|
0.04 ± 0.002 (36)
|
Vigna angularis (Willd.) Ohwi & H. Ohashi
|
114.1 ± 1.7 (35)
|
0.07 ± 0.004 (20)
|
Vigna radiata (L.) R. Wilczek
|
69.1 ± 1.0 (35)
|
0.05 ± 0.001 (15)
|
Phaseolus vulgaris L.4
|
334.8 ± 4.4 (738)
|
0.09 ± 0.001 (359)
|
Phaseolus coccineus L.
|
467.9 ± 8.3 (35)
|
0.09 ± 0.006 (15)
|
Cicereae
|
|
|
Cicer arietinum L.
|
189.6 ± 4.3 (35)
|
0.17 ± 0.004 (15)
|
Fabeae
|
|
|
Vicia faba L.4
|
613.9 ± 10.1 (210)
|
0.18 ± 0.004 (110)
|
Lens culinaris Medik.4
|
48.7 ± 1.3 (105)
|
0.05 ± 0.001 (45)
|
Lathyrus hirsutus L.
|
26.8 ± 0.6 (35)
|
0.15 ± 0.003 (12)
|
Lathyrus latifolius L.
|
57.3 ± 1.8 (35)
|
0.14 ± 0.006 (20)
|
Lathyrus odoratus L.
|
54.0 ± 0.9 (33)
|
0.12 ± 0.006 (16)
|
Lathyrus pratensis L.
|
15.0 ± 0.4 (35)
|
0.10 ± 0.002 (17)
|
Lathyrus sativus L.
|
248.7 ± 7.0 (35)
|
0.09 ± 0.007 (15)
|
Lathyrus tuberosus L.
|
37.0 ± 0.8 (70)
|
0.15 ± 0.003 (63)
|
Pisum sativum L.4
|
243.2 ± 1.6 (975)
|
0.09 ± 0.001 (489)
|
1Species’ and authors’ names are according to ILDIS (International Legume Database & Information Service) https://ildis.org/index.shtml [70]. Accessed 20 May 2020; 2Measured at room temperature; 3Measured under the dissection microscope; 4Cultivars/accessions were included in the measurements. |
Table 3
Egg-laying and larval development of the bean weevil (A. obtectus) on artificial seeds incorporated with bean seed coat powder
Type of pilule1
|
Number of eggs laid/female
(mean ± SE)
|
Dead L1 larvae outside (%)
|
Dead larvae inside (%)
|
Adults emerged (%)
|
2.5% seed coat
|
8.6 ± 2.1a
|
14.3
|
85.7
|
0
|
5.0% seed coat
|
4.5 ± 1.6a
|
28.9
|
71.1
|
0
|
10.0% seed coat
|
6.6 ± 1.2a
|
12.9
|
87.1
|
0
|
Control
|
7.6 ± 2.3a
|
2.7
|
2.7
|
94.6
|
1Pilules consisted of 80% cotyledon powder plus 20% water soluble potato starch powder. A portion of the latter was substituted with 2.5–10% seed coat powder. Control pilules contained only cotyledon and starch powders. The Brown-Forsythe test indicated homogeneity of variances of the number of eggs laid/female: F3,48=1.5344, p = 0.2176, and one-way ANOVA (F3,48=0.8445, p = 0.4763) and Scheffé post-hoc test were not significant at p < 0.05. |
Results of larval performance
Development of larvae. The major stages of larval performance were: (a) entering the seeds that was constrained by the thickness of seed coat, resulting in L1 mortality outside seeds, and (b) within-seed mortality of various developmental stages. Both were modulated by the intact or pierced status of seeds. The seed coats of H and ANH species were significantly thinner (H: 0.088 ± 0.002 mm, mean ± SE, N = 22; ANH: 0.103 ± 0.005 mm, N = 55) than those of NHs (0.126 ± 0.006 mm, N = 63) (Fig. 3, Kruskal-Wallis ANOVA: H2,140=17.48, p = 0.0002). L1 mortality outside intact seeds was significantly different among plant species groups (Fig. 3, Kruskal-Wallis ANOVA: H2,140=77.9, p < 0.001). It was lower for Hs (24.75%, 9.1–53.3 median and quartiles, N = 22) and ANH (93.3%, 80.5–100, N = 55) SCAs, in comparison with NHs (100%, N = 63). For H and ANH species Table 1 provides data of seed coat thickness, whereas ESM Tables 2 and 3 give similar information for NHs and SCAs of Hs and ANHs. Remarkably, seed coat substantiates a barrier even on its primary host, the beans (Ph. vulgaris), for the bean weevil. Contrary to the very low larval mortality inside seeds, there were substantial mortality outside and especially in case of intact seeds of bean cultivars (27.3%, 13.6–53.3, median, lower and upper quartiles, N = 21), and 4.4% (0-11.4) for pierced seeds. Similar values can be given for other legume species from which adults emerged, with the difference that the upper level of mortality usually reached 100% with intact seeds, with some exceptions such as Vigna unguiculata and V. angularis, where seed coats were extremely thin (Tables 1 and 2). A remarkable case is the runner beans (Ph. coccineus) with 0% (intact seed) and 2.2% (pierced seed) larval mortality outside seeds.
Table 2
Emergence and length of development of bean weevil (A. obtectus) adults and mortality of stages in host and acceptable non-host plant species. Where data were obtained in percentages, results were presented in medians and quartiles (numbers in smaller fonts). Glycine max comprises 17, Vigna unguiculata 2, Phaseolus vulgaris 21, Vicia faba 6, Lens culinaris 2, Pisum sativum 27 cultivars, and Lathyrus tuberosus 2 samples
Plant tribe and species1
|
Adult emergence (%) from seeds
|
Length of development (days)2
|
L1 mortality (%)
|
Larval-pupal mortality (%)
|
|
|
Outside seeds
|
Inside seeds
|
|
Intact
|
Pierced
|
Intact
|
Pierced
|
Intact
|
Pierced
|
Intact
|
Pierced
|
Genisteae
|
|
|
|
|
|
|
|
|
Lupius albus
|
0
|
3.5
|
|
|
100
|
18.8
|
0
|
77.7
|
Phaseoleae
|
|
|
|
|
|
|
|
|
Glycine max
|
0*
|
0
0-2.2
|
68–101
|
67–92
|
97.7
93.2–100
|
2.3
2.2–4.4
|
2.2
0-6.7
|
95.5
90.9–97.8
|
Lablab purpureus
|
0
|
7.0
|
|
|
100
|
10.3
|
0
|
82.7
|
Vigna unguiculata
|
65.2
39.5–90.9
|
93.3
88.9–97.7
|
30–40
|
28–51
|
32.5
6.8–58.1
|
3.3
0-6.7
|
2.3
2.3–2.3
|
3.4
2.3–4.4
|
V. angularis
|
11.4
|
63.6
|
51–154
|
41–68
|
34.1
|
4.6
|
54.6
|
31.8
|
V. radiata
|
0
|
86.7
|
|
30–40
|
100
|
2.2
|
0
|
11.1
|
Phaseolus vulgaris
|
72.7
46.7–86.4
|
93.2
86.1–97.6
|
31–59
|
31–86
|
27.3
13.6–53.3
|
4.4
0-11.4
|
0
0-2.2
|
2.2
0-2.3
|
P. coccineus
|
100
|
97.8
|
37–44
|
37–44
|
0
|
2.22
|
0
|
0
|
Cicereae
|
|
|
|
|
|
|
|
|
Cicer arietinum
|
2.2
|
97.8
|
|
33–58
|
97.8
|
2.2
|
0
|
0
|
Fabeae
|
|
|
|
|
|
|
|
|
Vicia faba
|
6.7
2.3–20.5
|
38.4
35.9–41.9
|
32–68
|
35–79
|
83.9
73.3–95.4
|
35.6
30.2–51.3
|
4.7
2.3–15.6
|
19.2
9.3–31.8
|
Lens culinaris
|
0
|
2.3
0-2.3
|
|
|
100
95.6–100
|
51.2
34.9–56.8
|
0
0-4.4
|
48.8
40.9–62.8
|
Lathyrus hirsutus
|
0
|
20.0
|
|
84
|
100
|
32.0
|
0
|
48.0
|
L. latifolius
|
0
|
4.2
|
|
85
|
100
|
87.5
|
0
|
8.3
|
L. odoratus
|
0
|
12.0
|
|
85
|
100
|
32.0
|
0
|
56.0
|
L. pratensis
|
0
|
46.7
|
|
84
|
100
|
6.7
|
0
|
46.7
|
L. sativus
|
57.8
|
75.6
|
33–90
|
33–47
|
40.0
|
11.1
|
2.2
|
13.3
|
L. tuberosus
|
0
|
13.6
0-28.9
|
|
40–42
|
100
|
28.0
26.7–59.1
|
0
|
44.4
27.3–72.0
|
Pisum sativum
|
0
0-4.4
|
20.5
8.9–34.9
|
46–101
|
40–162
|
93.2
83.7–95.6
|
7.3
4.5–18.9
|
4.6
2.2–7.7
|
62.5
46.3–85.0
|
1According to ILDIS (International Legume Database & Information Service) https://ildis.org/index.shtml [70]. Accessed 20 May 2020; 2From L1’s entering the seeds to adult emergence (min-max values). Empty cells mean missing data. *Although the median was zero, a very low number of adults emerged from intact seeds (mean ± SE: 1.12 ± 0.6, N = 17). |
Mortalities of various developmental stages inside intact seeds were substantially different from those of pierced ones. Furthermore, there were differences among host-types too (Kruskal-Wallis ANOVA: H2,140=28.39, p < 0.001). Table 2 provides results for H and ANH species, and further data are available in ESM Tables 2 and 3 for bean, pea and soybean varieties, as well as for NH species. For NHs mortality inside intact seeds was practically nil due to the inability of L1s to penetrate the seed coat, whereas for pierced seeds it was 75% (44–93%, median and quartiles). Many L1s first entered the seeds then left them again and died outside of starvation or by toxins taken up from the cotyledon. Seed testa frequently bore several shallow pits where L1 attempted to bore in (e.g. all Gleditsia japonica seeds, both intact and pierced, had such traces). It is worth remarking that ca. 10% of L1 entered the intact Gleditsia delavayi seeds. Inside the seeds many L1s died by various manners. Some perished on the surface of the cotyledon, others following preparation of tunnels of various lengths. Although it is well documented that additional A. obtectus larvae may enter through the hole made by a pioneer larva [21], the ratio of L1-made and artificial holes was 1.5:1 on the most preferred bean seeds (Ph. coccineus), i.e., most L1 larvae did not use the pre-prepared holes on this host. In some instances (Caragana, Onobrychis) L1s entered the seeds through the hilum. First instar larvae entering Gleditsia seeds through an artificial hole made at the embryo area invariably died within the embryo making 1–2 mm long tunnels. In cases where the cotyledons were soft (several Glycine cultivars/accessions and Caragana) larvae made longer paths before dying.
Development in artificial seeds. The experiment proved that besides being a physical barrier, seed coat also inhibited larval development at the lowest concentration incorporated into artificial seeds (Table 3).
Adult emergence. Significantly more adults emerged from H species/cultivars than either from ANH species/cultivars or NH species, and whether intact or pierced (Table 2). From intact seeds of Hs 75.3% (46.7–90.5), from ANHs 0% (0–5, medians and quartiles), and from NHs 0% adults emerged (Kruskal-Wallis ANOVA: H2,140=120.88, p < 0.001). From pierced seeds of Hs 93.3% (86.1–97.7 medians and quartiles), from ANHs 26.7% (8.9–48.8), and from NHs 0% adults emerged (Kruskal-Wallis ANOVA: H2,140=93.61, p < 0.001) (Fig. 1). Of the 62 legume species A. obtectus larvae developed into adults in 18 (29%) in four tribes (Table 2), however, the picture varied considerably concerning SCAs, and whether the seed coat was intact or not (ESM Table 3). Although adult emergence in Ph. vulgaris was generally high, at cultivar level it ranged between 51 and 100%. In Ph. coccineus all larvae developed into adults. Whereas cowpea (Vigna) species supported larval development to adults, soybean (Glycine) varieties and hyacinth bean (Lablab) did so only sporadically. This is also supported by the length of developmental time needed until adult emergence (Table 1). In some cases (G. max, V. angularis and P. sativum) it was two to three times longer in comparison with beans. Of the 27 cultivars of garden peas, adults emerged from 24 (88.9%), however, only from 13 of these if the testa was intact. Similar values for 17 G. max cultivars/accessions: adult emerged from six (35.3%), and only from four of these if having intact seed coat. A surprising feature is the asymmetric distribution of adult emergence within the Fabeae tribe. Whereas there was adult emergence from six Lathyrus species (one, L. sativus, from intact, the rest of them from pierced seeds), there was only one such case among vetches, the faba beans (Vicia faba). Both intact and pierced seeds yielded adults from faba beans, although it had a relatively thick testa. Although larval development proceeded in some NHs reaching as far as L3, no adult emerged from these (ESM Table 2).
In the 18 adult-yielding legume species five (28%) produced malformations and some adults could not leave the seed in 10 of these. Typical malformation was a substantial decrease in elytra width and length (the elytra became shorter and triangular in shape). There were 0.06% malformed adults in beans, 33.0% in peas, 31.6% in L. tuberosus, 6.67% in L. sativus, and 0.72% in V. faba.
Female preference and larval performance relation
Correlations. Table 4 provides the most important correlation coefficients referring to the overall relationship between plant traits and insect responses. Accordingly, only in the ANH group there was a significant positive relationship between preference and performance, i.e., the number of eggs laid and the adult emergence. The seed mass and the number of eggs laid positively, but not significantly, correlated only in Hs. The thicker the seed coat, the higher was the first instar larval mortality outside on ANH and NH seeds, but seed coat thickness did not affect L1 larval mortality on H seeds.
Table 4
Nonparametric (Kendall τ) correlation coefficients between seed traits and/or insect responses of the bean weevil (A. obtectus)
Correlations between seed traits and/or insect responses
|
H
|
ANH
|
NH
|
Seed mass vs. Number of eggs laid per female
|
0.2739
|
0.0189
|
0.0866
|
Seed coat thickness vs. L1 mortality outside seeds
|
-0.3414
|
0.2139
|
0.2264
|
Number of eggs laid per female vs. Adult emergence
|
0.1400
|
0.3088
|
-
|
H = hosts, ANH = acceptable non-hosts, NH = non-hosts. Bold numbers are significant correlations (p < 0.05). |
Results of logistic regressions. The logit-regression provided evidence that L1 mortality outside seeds was due to different seed coat thicknesses. Significantly (15:1) higher number of cases indicated < 50% L1 mortality outside seeds, if seed coat was thin (0.081 ± 0.001 mm, mean ± SE, N = 62), in comparison with thick seed coat (0.147 ± 0.005 mm, N = 78; Wald test: 8.165, df = 1, p = 0.0043; log-likelihood: -48.4233; goodness of fit χ2: 17.986, df = 1, p < 0.001). However, approaching the same hypothesis from another angle, i.e., if seed coat was assigned as ’penetrable’ or ‘impenetrable’, significantly higher number of L1 entered seeds with ‘penetrable’ seed coat (Wald test: 27.2694, df = 1, p < 0.001; log-likelihood: -81.4425; χ2 = 31.082, df = 1, p < 0.001). Here only those cases were taken into account where L1 larvae entered a seed then died immediately after this. This result is interesting, because ‘penetrable’ seed coat thickness was 0.0998 ± 0.004 mm (N = 64), whereas ‘impenetrable’ showed 0.119 ± 0.005 mm (N = 76), amounting only to 0.02 mm difference.
As for the second hypothesis, larval performance and adult emergence did depend on the ‘quality’ or ‘suitability’ of cotyledons of seeds. Significantly higher number of cases showed < 30% larval mortality inside seeds, if cotyledon was ‘suitable’ in comparison with ‘unsuitable’ cotyledon (Wald test: 27.5997, df = 1, p < 0.001; log-likelihood= -131.456; χ2 = 29.7737, df = 1, p < 0.001). The number of eggs on ‘suitable’ seeds was 26.25 ± 1.11 eggs/female (mean ± SE, N = 124) vs. on ’unsuitable’: 13.84 ± 0.72, N = 88. Logistic regression for adult emergence could not be performed, because one cell of the χ2 table contained zero. Nevertheless effect sizes could be computed.
Effect sizes and risk analyses. The first phase of the preference-performance relationship refers to the larval ability to overcome seed coat thickness. Odds ratio (OR) provided 19 times larger chance for L1 larvae having < 50% mortality, if they happened to bore in a seed with seed coat thickness of < 0.1 mm, and risk difference (RD) and risk ratio (RR) also strengthened the finding (Table 5). However, the regression coefficient (φ2) explained only 10% of variance of seed coat effect. On the other hand, when facing penetrable/impenetrable seed coat (results are not shown in table), L1 larvae got into the ‘penetrable’ seeds with ca. 3 times higher probability (RR = 3.01 ± 1.3, mean ± SE, CI95 = 1.9 & 4.8), but effect size for larvae to enter a seed was large (OR = 7.5 ± 1.5, CI95 = 3.5 & 16.1) in case of a penetrable testa, even if they died after the first bites from the cotyledon. Here, the regression coefficient explained a relatively high level (21%) of the variance.
Table 5
Chi2-table and risk effect sizes with 95% confidence intervals of bean weevil (A. obtectus) L1 mortality due to seed coat thickness on hosts and acceptable non-hosts (lumped together) vs. non-hosts
Dependent
|
Independent variable: Seed coat thickness
|
Effect size
|
variable
|
|
< 0.1 mm
|
> 0.1 mm
|
Marg. sums
|
|
|
L1
mortality
|
|
|
|
p< 50%3= 0.95
p> 50%4= 0.4917
|
First instar
|
< 50%1
|
19
|
1
|
20
|
RD5 = 0.4583; CI956 = 0.30 & 0.62
|
mortality
|
L1
|
|
|
|
RR7 = 1.9322; CI95 = 1.57 & 2.38
|
outside
|
mortality
|
|
|
|
OR8 = 19.7; CI95 = 2.55 & 152.4
|
seeds
|
> 50%2
|
59
|
61
|
120
|
logit d9 = 1.6441
|
|
Marginal
|
|
|
|
Pearson correlation (φ) = 0.3229
|
|
sums
|
78
|
62
|
140
|
Regression (φ2) = 0.1043
|
1Hosts and acceptable non-hosts; 2Non-hosts; 3Proportion of L1 died in case of hosts and acceptable non-hosts; 4Proportion of L1 died in the presence of non-hosts; 5Risk difference; 695% Confidence Interval; 7Risk ratio; 8Odds ratio; 9Standardized mean difference; χ2 for the table = 14.6, df = 1, p < 0.001. |
The second phase of the preference-performance relationship is manifested in the adult progeny production as related to the number of eggs laid per species. The chance of < 30% mortality inside seeds for larvae increased ca. 2 times (RR = 1.95 ± 1.13, CI95 = 1.52 & 2.49) in cotyledon suitable for reaching later developmental stages. The odds of outcome was high (OR = 4.85 ± 1.28, CI95 = 2.96 & 7.94, results are not shown in table). The chance for reaching adulthood in seeds where > 10% of adults emerged was ca. 4 times higher (RR = 3.82 ± 1.16, CI95 = 2.83 & 5.16), than in cotyledons where only 1–2 weevils could successfully terminated development. The odds for adulthood in suitable seeds were extremely high (OR = 523.42 ± 1.23, CI95 = 348.74 & 785.60) due to the asymmetry caused by the NH seeds (Table 6). The regression coefficient (φ2) explained a high portion (54%) of the variance.
Table 6
Chi2-table and risk effect sizes of bean weevil (A. obtectus) adult emergence from hosts and acceptable non-hosts (lumped together) vs. non-hosts
Dependent
|
Independent variable: Suitability (suit.) of cotyledon
|
Effect size
|
variable
|
|
More
suit.
|
Less suit.
|
Marg. sums
|
|
|
|
|
|
|
p> 10%3= 0.9946
p< 10%4= 0.2603
|
Adult
|
> 10%1
|
92
|
0
|
92
|
RD5 = 0.7343; CI956 = 0.69 & 0.77
|
emergence
|
|
|
|
|
RR7 = 3.8206; CI95 = 2.83 & 5.16
|
|
|
|
|
|
OR8 = 523.4; CI95 = 348.7 & 785.6
|
|
< 10%2
|
31
|
89
|
120
|
logit d9 = 3.4534
|
|
Marginal
|
|
|
|
Pearson correlation (φ) = 0.7373
|
|
sums
|
123
|
89
|
212
|
Regression (φ2) = 0.5437
|
1Hosts and acceptable non-hosts; 2Non-hosts; 3Proportion of adults emerged from hosts and acceptable non-hosts; 4Proportion of adults emerged from non-hosts; 5Risk difference; 695% Confidence Interval; 7Risk ratio; 8Odds ratio; 9Standardized mean difference; χ2 for the table = 117.61, df = 1, p < 0.001. |
As expected, there was an interaction between penetrable/impenetrable and intact/pierced states of seed coats: the joint effect (OR11 = 0.6718) was larger than the multiplied value (0.4820) of their respective effects (OR10 = 0.7179 and OR01 = 0.4820).
A model selection procedure among seed traits by GLZ logit regression to test whether seed coat thickness or suitability of cotyledon for development was more influential in affecting bean weevil responses showed that seed coat was overwhelmingly deterministic in the subsequent fate of larvae and, therefore in adult emergence too (all Wald statistics were significant).