1. Introduction
Lepidoptera stemborers, among the superfamilies of Noctuoidea and Pyraloidea, attack host plants belonging to the Poaceae, Cyperaceae, Typhaceae, and Juncaceae families with overlapping spatial and temporal distribution. Several studies report that, while more than 300 stemborer species infest wild plants, only 21 stemborer species attack cereal crops, mainly maize (
Zea mays L.), sorghum (
Sorghum bicolor L.), and millet (
Pennisetum glaucum (L.) R. Br.), in various parts of Africa [
1,
2,
3,
4,
5,
6,
7,
8]. A few, i.e., the noctuids
Busseola fusca Fuller and
Sesamia calamistis Hampson, the crambid
Chilo partellus Swinhoe, and the pyralid
Eldana saccharina (Walker), are cereal crops of economic importance [
9]. Yield losses vary with the region but generally range from 10% to 80% depending on infestation by the pest species and the crop growth stage [
10,
11].
Lepidopteran stemborers that attack maize are poly/oligo-phagous and feed on other cultivated and wild plants [
12,
13,
14,
15,
16,
17,
18]. In sub-Saharan Africa, cereal crops are mostly grown in small fields surrounded by land that is occupied by wild plants of lepidopteran stemborers. Some poly/oligo-phagous stemborers that are found on wild plants, such as
Chilo orichalcociliellus Strand and
Pirateolea piscator (Fletcher), are occasionally found on cultivated cereal crops [
13,
19]. However, they are more frequently found on cultivated crops; for example,
Busseola segeta (Bowden) infests 13–61% of maize fields in Western Kenya [
20].
In Kenya, the larvae of
B. fusca (an African maize stalk borer),
S. calamistis (a pink stem borer), and
C. partellus (a spotted stem borer) contribute up to 82% of maize yield losses [
9,
21,
22,
23]. In the context of a biological control program, the most commonly used parasitoids are the larval parasitoids, including
Cotesia flavipes Cameron and
Cotesia sesamiae (Cameron) (Hymenoptera: Braconidae), followed by the pupal parasitoids
Xanthopimpla stemmator Thunberg (Hymenoptera: Ichneumonidae) and
Pediobius furvus Gahan (Hymenoptera: Eulophidae), and then the tachinid
Siphona sp. [
19,
24]. Among these natural enemies, the larval parasitoids
C. flavipes and
C. sesamiae have been found to be the most efficient to control lepidopteran stemborers and, for example,
C. flavipes was used in a classical biological control program against
C. partellus in East Africa [
24,
25].
Cotesia sesamiae efficiently parasitized
S. calamistis and
B. fusca larvae [
9]. During cropping seasons, the maize lepidopteran stemborers
B. fusca,
S. calamistis, and
C. partellus and their associated parasitoids
C. flavipes and
C. sesamiae are more abundant in maize fields as compared to wild plants [
26,
27,
28,
29]; and perennation (the ability to survive from one cropping season to the next) by both
C. flavipes and
C. sesamiae occurs mainly in cultivated habitats [
28,
29]. This supposes that the parasitoids follow their lepidopteran maize stemborer hosts during non-cropping seasons, either in larvae feeding on wild plants surrounding maize fields [
13,
17,
18,
30] or in diapausing larvae in maize residues left in the maize field after harvest [
13,
31,
32,
33,
34,
35]. However, since the abundance of both lepidopteran stemborers and their parasitoids is much lower in wild plants compared to maize plants [
26,
29], the role of wild plants as a reservoir for cereal stemborers and their parasitoids is still controversial, particularly in agro-ecosystems with reduced wild habitat.
In this context, the objectives of the present study were to (i) compare the diversity and abundance of stemborers and their associated parasitoids between maize residues and wild plants identified as hosts for maize stemborers [
2,
3] during non-cropping seasons; (ii) study the relationships between maize stemborer species and their respective parasitoid abundances in either maize residues or wild plants during non-cropping seasons and in maize plants during subsequent cropping seasons; and (iii) determine the main potential niches that could harbour parasitoids after releasing the larval parasitoids
C. flavipes and
C. sesamiae in the studied fields.
4. Discussion
Although a higher diversity of stemborers and parasitoids in wild habitats as compared to cultivated habitats has already been well-reported in the literature [
2,
3,
7,
19,
26,
27,
28,
29,
52,
53,
54,
55,
56,
57], this study highlights for the first time a broader host range in wild habitats as compared to maize residue habitats. This variation in distribution amongst insect diversity between wild and cultivated habitats might be a consequence of anthropogenic changes in the ecology of the availability of food resources, constraints of natural enemies, and the evolution of competitive interactions [
58,
59,
60]. Diniz et al. [
61], who studied species richness of flower-head insects (Tephritidae: Diptera) in natural and cultivated habitats, have mentioned that anthropogenic alterations in the landscape determine the impoverishment of insect diversity in cultivated habitats. For stemborers’ parasitoids, Mailafiya et al. [
28] showed that parasitoid diversity was lower in locations where maize cultivation was practiced on a commercial scale and where intense grazing activities persist across seasons. Nevertheless, a higher insect diversity does not mean a higher insect abundance. In fact, the abundance of stemborers, which are pest of maize plants (i.e.,
B. fusca,
S. calamistis, and
C. partellus) and their associated parasitoids (i.e.,
C. flavipes and
C. sesamiae), was found to be higher in the maize residue habitat as compared to the wild plant habitat. It was previously found that both stemborers and parasitoids associated with pest stemborers are generally less abundant in wild habitats than in maize plants of cultivated habitats [
26,
29,
52]. Although natural habitats surrounding cereal crops serve as refugia for sustaining the diversity of both stemborers and parasitoids from adjacent cereal fields [
12,
19], the abundance of stemborers and associated parasitoids is very low in the wild as compared to cultivated fields [
26,
61]. A possible explanation is the low abundance of wild plant species surrounding our studied fields. In fact, when analyzing the data obtained by Mailafiya [
12] in other agro-ecological zones in Kenya, the correlation between the abundance of stemborers in the maize fields and in maize residues was greater for maize fields surrounded by a low diversity of wild plants (Kitale and Mtito Andei) than for those surrounded by a high diversity of wild plants (Kakamega and Muhaka) (see
Tables S1 and S2 and Figures S1 and S2). In addition, these findings indicate that this correlation of the stemborer species abundance in maize plants of cultivated fields with those in maize residues depend not only on the abundance of wild plant species in the agro-ecosystem but also on the abundance of wild plants that are suitable to maize stemborers, such as
Megathyrsus maximus and
Sorghum arundicaneum (wild sorghum), surrounding maize fields. Another possible explanation is the generally higher survival and growth rates of the stemborers, and, thus, their associated parasitoids, on cultivated plants as compared to wild plants [
52,
62,
63,
64,
65,
66,
67].
Overall, the role of wild plants surrounding cultivated areas in the carry-over of stemborer pests and their associated parasitoids during the non-cropping seasons is limited, suggesting that other niches, such as maize residues, might be also involved. It has been well-reported that maize residues left in the maize field after harvest constitute an important source of the maize pest stemborers that are involved in the carry-over of the insect pests on maize plants for the subsequent cropping season [
13,
14,
15,
31,
32,
33,
34,
35,
68].
In our study, the highest Morista–Horn similarity indexes of both maize stemborers and their associated parasitoids obtained in maize plants of the cultivated habitat with maize residues as compared with the wild habitat indicate that maize residues might constitute an important refugia source not only of the maize stemborers but also of their associated parasitoids. Analyzing the data on stemborer and parasitoid recoveries between wild and maize residue habitats during non-cropping seasons that were obtained by Mailafiya [
12] in Kitale and Mtito Andei and Kakamega and Muhaga confirm our results (see
Tables S3 and S4). It was even reported that
S. calamistis populations living in wild habitats differ from those living in cultivated habitats [
69], which compromises wild habitats as a refugia source of that species coming from maize plants in cultivated habitats. Although it was shown that
B. fusca infestation might originate from specimens coming from outside the maize fields, probably from quite a distance [
70], based on the insect abundance relationships between maize residues and maize plants, we cannot preclude maize residues from being the main reservoir of maize stemborers, particularly when wild plants surrounding maize fields are scarce. It is well-reported that
B. fusca, for example, survive the dry season as larvae diapausing into maize residues left in the field after harvest [
14,
15,
31,
32,
34,
68]. The high positive correlation between the abundance of stemborers in the maize fields and in maize residues obtained in our study reinforced the fact that maize residues might serve as the main reservoir of maize stemborers and, thus, as the main source of the carry-over of the maize pest for the next cropping season.
In addition, for each species, stemborers recovered from maize residues gave rise to a significantly greater percentage of females as compared to stemborers recovered in maize plants. This is in accordance with Gebre-Amlak [
71], who reported that the first generation of
B. fusca coming from diapause larvae found earlier in the cropping season gave more females than males compared to further generations. These seasonally dependent sex ratio variations might be due to either climatic and environmental factors or intrinsic factors of the insect to ensure the perennity of its species by a female-biased sex ratio distortion when the conditions became unfavorable. Kageyama et al. [
72], studying the occurrence of feminizing bacteria in an insect by a female-biased sex ratio in
Ostrinia furnacalis (Lepidoptera: Crambidae), the Asian corn borer, concluded that, in the sex determination systems in lepidopteran insects, chromosomal males are feminized by a cytoplasmic agent, most probably parasitic bacteria, according to the conditions. This phenomenon was confirmed in other insect species [
73,
74,
75,
76,
77] and might be explored in lepidopteran stemborers in relation to the habitat.
In addition, the absence of both
C. flavipes and
C. sesamiae in the field before any release could be due to: (i) the frequent use of pesticides; (ii) the systematic use of maize residues for animal feed during dry seasons; or (iii) the climate change adaptation of both parasitoids and hosts to dry seasons, which, in the last five years, have been particularly long [
36]. In contrast, after and during the releases, although the parasitism rates were low (which normally occurs after the first parasitoid release [
45] and according to the season, the year, and the locality [
78,
79]), the highest recovery of
C. flavipes and
C. sesamiae obtained in maize fields during the subsequent cropping seasons indicate a possible establishment of the parasitoids released in these areas. The successful establishment of
C. flavipes after release has already been observed in different countries, including coastal Kenya [
24,
26,
45,
55,
80]. In addition, the fact that, similarly to maize stemborers,
C. flavipes and
C. sesamiae were mostly recovered after and during the releases in maize residues during the non-cropping seasons confirmed those residues as being the main reservoir of maize stemborer parasitoids during dry periods. This aspect is important to consider in the context of biological control. In fact, it has been recommended that the maize residues (observed to be an important reservoir of maize stemborers) be burned [
14,
15,
31,
32,
66] in order to diminish the risk of maize infestation by stemborers for the subsequent cropping seasons [
5]. This management measure should not be adopted in areas where
C. flavipes and
C. sesamiae have been released or where the wild habitat has been drastically reduced. In those contexts, maize residues might ensure the perennity of the parasitoids during dry seasons. Considering that: (i)
C. flavipes and
C. sesamiae were found to be rare or absent in all habitats prior to release; (ii) parasitism by
C. flavipes is generally low or absent in the years after biological control release [
79]; and (iii) maize residues are also the main reservoir of parasitoids during dry periods [
12] (see
Tables S2 and S4), maize residues may also represent the main sources of stemborer parasitoids during non-cropping seasons. This suggests that maintaining residues will promote the parasitism of stemborers. However, to determine whether a buildup of the parasitoid population might occur over time, a further study might be conducted on the comparable effect of infestation and parasitism over a longer time period to confirm the influence of maize residues on both infestation and parasitoid presence. In addition, some wild plants, such as the wild sorghum
S. arundinaceum, support a high survivorship of parasitized stemborers and, therefore, a relatively high performance of their larval parasitoids [
64]. The maintenance of these wild plants is also vital for the survival and, thus, the perenity of
C. flavipes and
C. sesamiae in the field.