Bioacoustics of Megatibicen pronotalis

REMARKS. The historical taxonomy of the species and proper application of the name with priority was reviewed by Hill & Marshall (2009) with additional updates to the generic placement by Hill et al . (2015) , Sanborn & Heath (2016 , 2017a ), and Lee (2016) . Davis (1938) identified examples from the Dakotas, Iowa , Nebraska , and Oklahoma that were smaller and had a large piceous mark on the dorsomedial pronotal collar. Although this mark is found in other populations, it is found in a minority of specimens ( Davis (1938) mentions about 2% of a population of specimens from Louisiana ). He also mentions that the mesonotum is considerably smaller and the mesonotal mark in these specimens is smaller than the typical form. Finally, Davis (1938) states the new variety is more similar in appearance to Megatibicen cultriformis ( Davis, 1915 ) than the typical form of the species ( M. pronotalis walkeri rev. stat. at the time). Davis (1938) identified morphological differences in his proposed variety beyond a simple color variation that Kratzer (2024) mistakenly identifies as the single variable trait used to form the subspecies. There are data already published illustrating there are morphological differences between the subspecies. Statistically significant differences in body mass (t=2.1346, d.f.=17, P=0.0476), wing length (t=7.9020, d.f.=6, P=0.0002), and wing span (t=5.0241, d.f.=6, P<0.0024) can be shown in the already published data ( Sanborn et al . 2021 ). Further analyses ( Tables 1 and 2 ) provided below have identified additional structures that differ between the previously assigned subspecies as demonstrated here with three measurements ( Table 1 ). Morphological Variable Dakota form Common form Western form Common + Western forms Statistic (d.f.) Probability (n=2) (n=6) (n=4) (n=10) Body length (mm) 34.85 ± 1.3435 37.09 ± 0.2853 40.86 ± 1.1514 38.60 ± 2.0685 F=41.8545 (2, 9) P<0.0001 t=2.4112 (10) P=0.0366 Mesonotum width (mm) 13.05 ± 0.9989 14.05 ± 0.3633 15.64 ± 0.5543 14.69 ± 0.9208 F=18.4501 (2, 9) P<0.0007 t=2.2816 (10) P=0.0457 Mesonotum length (mm) 9.23 ± 0.1061 10.15 ± 0.1673 11.26 ± 0.4573 10.58 ± 0.6273 F=33.7091 (2, 9) P<0.0001 t=2.9348 (10) P=0.0149 Access to male specimens from South Dakota (n=2), Texas (n=4), and Missouri , Arkansas and Mississippi (n=6) in the author’s collection provided the opportunity to test for potential differences between, respectively, the Dakota form (equivalent to M. pronotalis pronotalis rev. stat. ), the western form (equivalent to the southwestern range of M. pronotalis walkeri rev. stat. considered M. pronotalis hesperius n. ssp. here ), and the common form (equivalent to the majority of M. pronotalis walkeri rev. stat. ) of M. pronotalis identified by Kratzer (2024) ( Fig. 1 , Table 1 ). Body length, mesonotum width, and mesonotum length were measured first since Davis (1938) identified the body length and the mesonotum as being smaller in his new variety. It is not clear if Davis (1938) was referencing mesonotum width or length so both were measured and compared here. If the Dakota form is compared to a combination of the western and common forms (the previously defined subspecies), there are statistically significant differences in body length (t=2.4112, d.f.=10, P=0.0366), mesonotum width (t=2.2816, d.f.=10, P=0.0457), and mesonotum length (t=2.9348, d.f.=10, P=0.0149). If the three forms are considered to be independent populations, eight of the nine possible pairings are significantly different statistically. ANOVA analyses of body length (F=41.8545, d.f.=2, 9, P=2.767x10 -5 ), mesonotum width (F=18.4501, d.f.=2, 9, P=6.545x10 -4 ), and mesonotum length (F=33.7091, d.f.=2, 9, P=6.602x10 -5 ) show statistically highly significant differences between the populations. Tukey Kramer tests, used to determine which populations differ from one another, are significant in all but one population pair for mesonotum length. Body length differs statistically in all species pairs (Dakota vs. common Q=4.6821, P=0.02231; Dakota vs. western Q=11.84, P=4.089x10 -5 ; common vs. western Q=9.9625, P=1.592x10 -4 ). Mesonotum width differs statistically in two of the three species pairs (Dakota vs. common Q=3.2464, P=0.1078; Dakota vs. western Q=7.9197, P=8.722x10 -4 ; common vs. western Q=6.5189, P=0.003269). Finally, mesonotum length differs in all species pairs (Dakota vs. common Q=5.4466, P=0.009806; Dakota vs. western Q=11.1029, P=6.829x10 -5 ; common vs. western Q=8.0067, P=8.070x10 -4 ). Although some of the species pairs differ in their significance if a non-parametric Kruskal Wallace test is performed due to the low sample sizes, all variables still show statistically significant differences in body length (P=0.009898), mesonotum width (P=0.01312), and mesonotum length (P=0.009739). The Dunn’s test shows that the western form is significantly different from both the Dakota and Common forms in all variables (P< 0.032 in all cases). These limited sample size populations (which make it more difficult to show significant differences) statistically support the presence of three distinct populations using only three morphological features with very high probabilities in most comparisons. Kratzer (2024) appears to have identified a third subspecies in her western form rather than providing evidence that the two previously identified subspecies should be synonymized. Kratzer (2024) also attempts to show the songs of the two subspecies are the same as another factor to support synonymy of the subspecies. She cites the website of Marshall & Hill (https://insectsingers.com) as the source of songs to show similarities between the two subspecies but provides no information as to how the song parameters were measured or the sensitivity of the analysis equipment. The overlap in the song frequency cited would be expected based on the similarities in body size of the subspecies and cicada call frequency being determined by body size (Bennet-Clark & Young 1996). The smaller M. pronotalis pronotalis rev. stat. is also reported to have the higher frequency call ( Kratzer 2024 ) as would be predicted based on its smaller size, but the specific peak frequency is not provided. More concerning is the sample size for the analysis is one song for each subspecies. The 2.2 or 2.4 syllables per second reported by Krater (2024) may actually be different if a larger sample size is measured and these values are consistent or these represent extremes of non-overlapping populations. Songs of cicada subspecies have also been shown not to differ (e.g. Puissant & Gurcel 2023). In addition, the website and Hill et al . (2015) state that the songs of both subspecies are identical to Megatibicen dealbatus ( Davis, 1915 ) so that if we follow the logic of Kratzer (2024) that the two subspecies of M. pronotalis and the species M. dealbatus are the same because their songs are the same, then M. dealbatus would also be a synonym since the songs of all three taxa are identical. A compelling piece of evidence to support synonymy or the formation of three subspecies is the genetic analysis of Hill et al . (2015) that included both subspecies described at the time and examples of the three forms proposed by Kratzer (2024) . The single example of M. pronotalis pronotalis rev. stat. in the summary genetic analysis was found within the clade of three examples of M. pronotalis walkeri rev. stat. ( Hill et al . 2015 ). The specimen from Texas (identified as M. pronotalis walkeri ) is the first to diverge in the clade (which was collected near the southern end of the form distribution and was collected one county away from the specimens measured here), followed by the specimen from South Dakota ( M. pronotalis pronotalis rev. stat. ) and, finally, a branch with the specimens from Florida and Lousiana ( M. pronotalis walkeri rev. stat. ). This mixed grouping of subspecies in a single clade would normally be sufficient to synonymize the species but the differences in divergence of the Texas , South Dakota and two specimens from Louisiana and Florida are consistent in all genetic analyses even when the branching point positions change within clades. In fact, the Texas specimen and the South Dakota specimens each branch on different clAdes from the LoUIsIAnA And FlorIdA specImens when onlY the nUcleAr EF-1α gene Is Used to form A clAdogrAm ( Hill et al . 2015 , Fig. 4 ). The Texas specimen is always separated from the Louisiana and Florida specimens even though they were considered to be the same subspecies at the time. Kratzer (2024) identifies three distinct forms within the synonymied species that correspond to the branches of the different source populations within the clades of Hill et al . (2015) . The consistent alignment of the genetic analyses with the statistically significant morphological differences found between the three populations supports the contention that there are three subspecies rather than a single highly variable species that was formed when Kratzer (2024) synonymized the two available subspecies. Color variability found in iNaturalist photographs should not be considered a primary reason to synonymize the subspecies since Davis (1938) described this variability and its geographic distribution while demonstrating morphological differences in the populations. Even if parts of the different phenotypic populations are sympatric, there is no evidence that the different phenotypes interbreed and produce viable hybrids. What Kratzer (2024) did not address are the morphological differences described by Davis (1938) , confirmed in another study ( Sanborn et al . 2021 ), and further confirmed here. The difference in song frequency described by Kratzer (2024) supports the difference in body size between the subspecies but there are insufficient data to show the songs are the same in the subspecies or the three forms Kratzer (2024) proposes for the species. Several taxa that Davis originally described as varieties were later shown to be distinct species based on differences in morphology, song, ecology, and distribution ( Sanborn & Phillips 2001 , 2010 , 2011). Therefore, the synonymy of Megatibicen pronotalis pronotalis ( Davis, 1938 ) rev. stat. and Megatibicen pronotalis walkeri ( Metcalf, 1955 ) rev. stat. is not supported and the subspecies re-established here based on the consistent differences in the clade placement of the three populations in the genetic analyses of Hill et al . (2015) and the statistically significant differences in the morphological measurements ( Table 1 ) provided for the different populations above.A third subspecies is described here for the “Western” form of Kratzer (2024) based on the genetic and morphological differences described previously. I have chosen to describe a subspecies since the population had previously been considered part of a known subspecies. Further data, particularly analyses of the songs from the three subspecies, will help determine if they are subspecies or closely related species.