Introduction

The taxonomic status of members of the Melanoplus packardii group (Orthoptera: Acrididae) has been changed numerous times. Melanoplus packardii packardii Scudder was described in 1897 (Scudder 1897) and its taxonomic status has not shifted since (Pfadt 2002, Capinera et al. 2004). Melanoplus foedus Scudder was described in 1897 (Scudder 1897), but was considered a subspecies of M. packardii for five years (Blatchley 1920, Hebard 1925). It is currently recognized as a full species (Vickery & Kevan 1985, Pfadt 2002, Capinera et al. 2004). Melanoplus packardii brooksi Vickery was described from Canada in 1979 (Vickery 1979) and it remains recognized as a subspecies of M. packardii. Melanoplus fluviatilis Bruner was described as a species in 1897 (Bruner 1897), but without an extensive description. It was formally described in 1920 (Blatchley 1920), but was later placed as a subspecies of M. foedus (Hebard 1931). Its taxonomic status has remained unchanged since 1931 (Helfer 1987, Kirk &Bomar 2005). Melanoplus foedus isleyi Hebard was described as a subspecies of M. foedus and remains as a subspecies under current classification (Hebard 1936a, Capinera et al. 2004). Finally, Melanoplus stonei Rehn was described in 1904 (Rehn 1904) and has since been recognized as a subspecies of M. foedus (Cantrall 1968) or M. packardii (Hebard 1928), but is now considered distinct [Hebard 1932 (1931), Vickery & Kevan 1985, Kirk &Bomar 2005].

In summary, the current literature recognizes the following species and subspecies in the packardii group: Melanoplus packardii, Melanoplus foedus foedus, Melanoplus foedus fluviatilis, Melanoplus foedus isleyi, and Melanoplus stonei (Eades & Otte 2009). The ranges of M. foedus foedus and M. foedus fluviatilis overlap across a large portion of the United States, and yet no apparent hybrids have been mentioned in the literature. In addition, the range of M. foedus isleyi appears to overlap the range of M. foedus fluviatilis over a smaller area (Eades & Otte 2009).

All forms within this group are very difficult to differentiate because few morphological characters consistently differ between them and even the genitalic differences are few. The cerci of the males of all forms are virtually identical; however, the aedeagus of M. packardii differs from that of M. foedus (Brooks 1958).

Chapco et al. (1999) conducted mitochondrial DNA analyses on several North American Melanoplus species and determined that M. packardii and M. foedus were distinct, but closely related, species. However, genetic analysis of members of the group has also been inconclusive and a later study by Chapco and Litzenberger (2002) determined that M. packardii and M. foedus had some genetic overlap among individuals. Together, these results suggest that these species either represent a variable single species or they represent a very recent evolutionary radiation with hybridization possible. More detailed studies on the genetic relationships between described species and subspecies within the M. packardii group are needed to clarify these relationships.

Collections of large series from this group across Nebraska from 2005 to 2007 show trends that raise questions on the taxonomic status of forms currently recognized as subspecies of M. foedus and for M. packardii. This study scores morphological traits and uses molecular analyses to test three hypotheses: 1) Melanoplus foedus and M. packardii are distinct species; 2) Melanoplus foedus fluviatilis represents a full species; and 3) local hybridization occurs between M. packardii and M. foedus.

Methods

Large series of the forms M. packardii, M. foedus foedus, and M. foedus fluviatilis were collected from across Nebraska, as well as from a single site in South Dakota, during the summers of 2005–2007. Specimens were identified to species using a variety of resources, including Bruner (1897), Scudder (1897), Helfer (1987), and Pfadt (2002).

All specimens used for morphological studies were then pinned and placed in the collection at USDA-APHIS in Lincoln, Nebraska. Specimens representing potential hybrids between M. foedus foedus and M. packardii were grouped with the species or subspecies with which they shared the greatest similarity. Genetic examination was then used to determine the identity of these possible hybrids (see below). A total of 151 M. foedus fluviatilis, 265 M. foedus foedus, and 133 M. packardii were examined morphologically (Table 1).

Table 1.

Number of specimens of each Melanoplus taxon by county of collection examined in morphometric study.

Nonmeasurement data included species determination, date, county, locality, sex, hind tibial color, postocular bar development and predominant color, consistency of dorsal pronotal color (solid or striped), predominant dorsal pronotal color, and secondary pronotal color. Measurement data were obtained for tegmen length (from lowest insertion along lower pronotum to tip), narrowest dorsal distance between eyes, and length of male furculae. Measurements of tegmen length were done using a vernier caliper and estimated to the nearest 0.1 mm, while measurements of the intereye distance and length of furculae were taken with an optical micrometer (Scope: Olympus SZ-STS, 2.5x × 10x = 25x; Micrometer: Olympus AX0001 OB-M, 1/100 mm). Melanoplus f. fluviatilis and M. f. foedus were separated based on the characters listed in Table 2.

Characters were then analyzed using PROC GLIMMIX (SAS Institute Inc. 2006). Measurement data (tegmen length, distance between eyes, and male furcula length) were analyzed under a normal distribution while categorical variables (tibial color, postocular bar development, and postocular color) were analyzed under a multinomial distribution. Because of distinct differences in size between sexes, males and females were compared separately for measurement variables.

Examination of Aedeagi.—Aedeagi from three males of specimens identified as M. foedus fluviatilis, M. foedus foedus, and M. packardii were examined for structural differences. In each case, the terminal part of the abdomen of three males was severed, intestinal contents removed, and the remaining structure soaked in a solution of KOH for approximately 24 h. After 24 h, these were transferred to 70% ethanol and the aedeagus then removed. Aedeagi were cleaned under a dissecting scope to remove bits of connective tissue. They were placed in 70% ethanol until examination. Photographs of the aedeagi were taken through a dissecting scope, and drawn using a camera lucida. A comparison was made of the general structure of the terminal end of the aedeagus, especially in regard to paramere structure.

Genetic Analyses

DNA Isolation and Quantification.—DNA was isolated from the hind legs of 9 individuals of each form, following a modified Doyle and Doyle (1987) CTAB extraction protocol. Pelleted DNA was suspended in 50 µl 1xTE buffer (10 mM Tris-HCL, 0.1 mM EDTA). The locations, dates of collection, and the number of specimens used in the genetic analyses are presented in Table 3.

AFLP-PCR.—Genetic variation was analyzed using a modified AFLP (amplified fragment length polymorphism) procedure based on Vos et al. (1995).

Template preparation. —Approximately 7 µl of 150 ng/µl DNA template was incubated with 0.0625 µl EcoRI, 0.125 µl MseI (New England Biolabs), 1.25 µl One-Phor-All buffer (GE Healthcare), 0.125 µl bovine serum albumin (New England Biolabs), and nanopure water for total volume of 12.5 µl at 37^°C for 2.5 h in a PTC-200 Peltier Thermal Cycler (MJ Research, Inc., Waltham, MA). The resulting fragments were then incubated at 25 ^°C for 8 h with a ligation mixture of 0.15 µl T4 DNA ligase, 10x T4 DNA ligase buffer (New England Biolabs, Foster City, CA), 0.5 µl EcoRI adapter, 0.5 µl MseI adapter (Operon Technologies), and 3.35 µl nanopure water. A 1: 10 dilution was then performed on the ligation product using Ix TE buffer.

Table 2.

Characters for differentiating M. foedus fluviatilis and M. foedus foedus.

Preamplification.—1.25 µl of the ligation mixture was incubated with 10 µl Preamplification Primer Mix II (Invitrogen), 1.25 µl 10x PCR buffer II, 0.75 µl MgCl₂, and 0.25 µl Amplitag DNA polymerase (Applied Biosystems). The PCR program consisted of 20 cycles (30 s at 94^°C, 1 min at 56^°C, 1 min at 72^°C). A 1:20 dilution with nanopure water was performed on the product.

Selective amplification.—Reaction volumes containing 4.1 µl nanopure water, 1.2 µl 10x PCR buffer II, 0.72 µl MgCl₂, 0.08 µl Amplitag DNA polymerase (Applied Biosystems), 2.0 µl MseI primer (LI-COR), 0.4 µl EcoRI IRD-700 labeled primer (LI-COR), and 2.0 µl of the preamplification template were amplified via PCR. Primers were screened and chosen based on the number and clarity of bands produced. The PCR program consisted of one cycle (30 s at 94°C, 30 s at 65°C, 1 min at 72°C), 12 cycles (30 s at 94°C, 1 min at 72^°C), and 23 cycles (30 s at 94^°C, 30 s at 56^°C, 1 min at 72^°C). The reaction was stopped by adding 2.5 µl stop solution (LI-COR). The product was then denatured for 1 min at 94^°C and stored at-20^°C.

Fig. 1.

County-level distribution of specimens collected for morphometric analyses across Nebraska. Light gray indicates Melanoplus foedus, medium gray indicates M. fluviatilis, and dark gray indicates M. packardii. Counties with checkered pattern had more than one species analyzed.

Data scoring and analysis.—One-microliter samples were electrophoresed through a KBplus 6.5% Polyacrylamide gel (LI-COR) and the bands detected via infrared florescence, using a laser scanning machine (LI-COR Model 4200S-2). An IRD-700 labeled 50–700 bp size standard was used to estimate fragment size. Sixty-two markers were selected based on clarity and bands were scored using the program SAGA MX 3.2 (LI-COR). The data were converted to matrix form for further analysis, with a 1 indicating band presence and a 0 indicating absence. Data were analyzed using PAUP 4.0b 10 (Swofford 2001 ). Distance analysis was performed using neighbor joining and minimum evolution, while unweighted maximum parsimony was performed using a heuristic search. Bootstrap analyses of 1000 replicates were performed to assess clade support. Arphia xanthoptera (Burmeister) served as an outgroup taxon. One specimen of the outgroup and a total of 34 specimens of the selected taxa were used in the analyses. Two specimens of the selected taxa did not demonstrate markers distinct enough for analysis.

Results

Distribution. —M. foedus foedus and M. foedus fluviatilis were only found to co-occur at a single site; apparent intergrades were extremely rare and only found at that site. Elsewhere in Nebraska each of these two forms occupied distinct habitats, with M. f. foedus occurring in dry sandy uplands and M. f. fluviatilis in open sandy woodlands and the upper reaches of sandbars along major rivers (Fig. 1).

Table 3.

Collection information for specimens used in genetic analyses.

Extensive sampling showed M. foedus and M. packardii co-occur only rarely in Nebraska. However, in some areas of co-occurrence apparent hybrids between M. foedus and M. packardii were collected fairly commonly, and at these sites, “hybrid” forms tended to predominate over either of the two parent species, making the specie difficult to distinguish. At one site, in Keith County, Nebraska, both species co-occurred in large numbers and no apparent hybrids were observed. This suggests that these two species may be hybridizing in some areas and that the hybridization might be at least partially driven by low abundance of the two forms at these sites.

Morphological examination.—Once separated based on original species descriptions (see Eades and Otte 2009), each of the three forms were found to differ statistically in several characters. All three forms differed significantly in tibial color frequency (p <0.0001). Tibial color of M. foedus was nearly always red to pinkish red and that of M. packardii was most often blue (Fig. 2). The tibial color of M.f. fluviatilis was variable, being red, purple, or blue, or more rarely, pallid (Fig. 2). With the exception of a single specimen from Lake McConaughy, Keith County, Nebraska, all M. foedus fluviatilis had the dorsal pronotum solidly colored (usually brown), and this single specimen may be an intergrade with M. foedusfoedus. All M. foedus foedus and M. packardii had the dorsal pronotum distinctly striped. While M. foedus foedus and M. packardii were generally similar in having a striped dorsal pronotum, the color of the dark stripes tended more strongly toward red-brown in specimens identified as M. foedus foedus and dark olive-brown in M. packardii.

Fig. 2.

Proportions hind tibial colors among the 3 species. Number of specimens examined: M. foedus fluviatilis (n=150), M. foedus foedus (n=264), and M. packardii (n=133).

Hind femur and tegmen lengths were similar within sexes for all three forms (Table 4). Both sexes of M. f. fluviatilis and M. packardii differed from M. f. foedus in the minimum distance between the eyes, but were similar to each other (Tables 4 and 5). Males of Melanoplus f. fluviatilis and M. packardii differed from M. f. foedus in the length of the male furculae (Table 5).

Examination of the aedeagi of males showed strong differences in the shape and angle of the basal ring, as well as in the shape and length/width ratio of the parameres among adeagi of M. packardii and M. foedus. There were no distinct or consistent differences between M. foedus fluviatilis and M. foedus foedus in aedeagal structure. Although the middle tooth of the inner side of the primary ventral paramere appeared to be slightly more pronounced in M. f. foedus, this trait was not consistent (Fig. 3).

Genetic analyses.—A total of 62 characters, both monomorphic and polymorphic, were used in this study. Fifty-two of the 62 characters were parsimony-informative. A large number of polymorphic bands were found both between and within species. Neither distance nor parsimony analysis resolve this species complex (Figs 4, 5), suggesting that there may be frequent hybridization. It is unclear whether each of these forms is a distinct species or if hybridization and resultant introgression occurs. These results are similar to those obtained by Chapeo and Litzenberger (2002) when they analyzed numerous species in the genus Melanoplus using mitochondrial DNA. These authors were also unable to distinguish between M. packardii and M. foedus.

Table 4.

Means and standard errors of measurement data (mm) from analyzed specimens.

Discussion

As anticipated based on descriptions of the forms, Melanoplus f. foedus, M. f. fluviatilis, and M. packardii each display morphometric differences. Compared to M.f. foedus, M.f. fluviatilis is distinguished by having a solid color on the dorsal pronotum and usually having large, dark postocular bars. The minimum distance between the eyes was significantly different between M. f. foedus and M. f. fluviatilis, despite the fact that tegmen lengths did not differ. The structure of the aedeagus did not differ between the subspecies, but differed between M. foedus and M. packardii. Hind tibial color could also be used to differentiate M. f. foedus and M. packardii in Nebraska approximately 90% of the time. M. f. fluviatilis appears to be intermediate in this character, with individuals exhibiting broad variation in color. While M. stonei was not analyzed in this study, all specimens collected by the senior author from northwestern Ontario, Canada have red hind tibiae. To further complicate identification within this group, occasional specimens resembling M. foedus isleyi occur in Nebraska. This subspecies is differentiated by poorly developed postocular bars and pallid hind tibiae.

Examination of the male aedeagus revealed that in M. foedus it is clearly and consistently distinct from that of M. packardii in several characteristics. However, the aedeagus of M. f. fluviatilis is virtually indistinguishable from that of M. f. foedus. The shape and angle of the basal ring as well as the form of the parameres appear to be identical (Fig. 3).

Table 5.

Species and subspecies comparisons of tegmen length, minimum distance between eyes, and male furcula length. Significant differences (>0.05) are indicated by an asterisk and bold type.

The lack of any distinct clades in the genetic analyses (Figs 4, 5) supports M. foedus foedus and M. foedus fluviatilis as being the same species and that frequent interbreeding or hybridization occurs between these taxa and M. packardii. Our results are similar to those obtained by Chapeo and Litzenberger (2002) when they analyzed numerous species in the genus Melanoplus using mitochondrial DNA and were unable to distinguish between M. packardii and M. foedus based on genetic criteria.

Morphological differences in the absence of clear genetic differences suggest several possibilities: 1)M. foedus and M. packardii are environmentally induced forms of the same species; 2) they are species that were previously separated but have retained the ability to interbreed where populations meet; 3) or they are subspecies in the process of becoming full species. Although our data do not allow resolution of these or other possibilities, the forms in Nebraska occur most often in distinct habitats.

In Nebraska, M.f. foedus occurs in upland sandy areas with sparse vegetation and is especially common in the Nebraska Sandhills. Melanoplus f. fluviatilis occurs in lowland areas near rivers, especially along the higher areas of sandbars and in open woodland where there is more growth of weedy vegetation. Melanoplus packardii is most common in areas with somewhat sparse vegetation on generally heavy soils such as clay or loess. Populations of suspected hybrids between M. foedus and M. packardii have been found mostly in sparsely vegetated areas on heterogeneous soils, especially in areas where clays are mixed with small stones or gravel. These observations suggest that each form may differ in soil and/or plant preference, but may hybridize where multiple habitats mix.

In areas with heterogeneous soils, apparent hybrids between M. foedus and M. packardii tend to predominate over either parent species, but all forms tend to occur at low to moderate densities. At one site in Nebraska (Keith County, ca 1.9 km east of Roscoe off Highway 30), both M. foedus and M. packardii are abundant, but no apparent hybrids were found. At this site, it appears that the forms are capable of differentiating between each other. Together, these observations may lend evidence to male choice driving hybridization, as males of some insect species are known to mate with females of a closely related species if females of the same species are scarce (Platt et al. 1978).

Fig. 3.

Dorsal view of parameres and basal ring of aedeagi of M. foedus fluviatilis (top), M. foedus foedus (middle), and M. packardii (bottom).

The lack of genetic differentiation suggests that M. foedus and M. packardii are actually one species. However, we feel that differences in aedeagal structure and habitat occurrence should be recognized and that the species designation should remain until additional genetic or behavioral tests can be performed. The lack of resolution in the genetic tests could be the result of occasional hybridization.

While we conclude that M. packardii and M. foedus are distinct morphologically, including in male genitalia, we suggest that M.f. fluviatilis should be synonymized under M. foedus. Further research is needed to verify our conclusions, but we suspect that M. f. fluviatilis is an environmentally induced variant of M. foedus and does not qualify as a valid subspecies based on distribution. Both forms can be found throughout Nebraska in suitable habitat. Thus, one could feasibly consider them ecotypes; however, some of the external traits may be indiscrete or subject to environmental influence (Shelford 1917, Otte and Williams 1972). Such environmental factors have been shown to exert a strong influence on adult coloration in other grasshopper species (Otte & Williams 1972). Furthermore, collections of a related species, Melanoplus angustipennis (Dodge), have shown darker individuals more predominant in partially wooded riverine areas compared to in adjacent uplands (although not as dark as M. foedus fluviatilis), sometimes with this change noted over a distance of less than 0.2 km (M.L. Brust, unpub.).

Fig. 4.

Neighbor-joining tree. 1000 replicate bootstrap. Bootstrap values (>50) shown above branches. Letter abbreviations represent species and numbers represent individuals: OG = outgroup (Arphia xanthoptera), FL = Melanoplus foedus fluviatilis, FO = M. foedus foedus, FxP = M. foedus foedus × packardii?, PA = M. packardii.

Fig. 5.

Unweighted parsimony tree. 1000 replicate bootstrap. Bootstrap values (>50) shown above branches. Letter abbreviations represent species and numbers represent individuals: OG = outgroup (Arphia xanthoptera), FL = Melanoplus foedus fluviatilis, FO = M. foedus foedus, FxP = M. foedus foedus × packardii?, PA = M. packardii.