Necrophila (Calosilpha)

Růžička, Jan, Qubaiová, Jarin, Nishikawa, Masaaki & Schneider, Jan, 2015, Revision of Palearctic and Oriental Necrophila Kirby et Spence, part 3: subgenus Calosilpha Portevin (Coleoptera: Silphidae: Silphinae), Zootaxa 4013 (4), pp. 451-502 : 472-475

publication ID

https://doi.org/ 10.11646/zootaxa.4013.4.1

publication LSID

lsid:zoobank.org:pub:FE1C6E7B-1FFE-401B-928D-4900064068BE

DOI

https://doi.org/10.5281/zenodo.5686447

persistent identifier

https://treatment.plazi.org/id/039787D7-560E-FF8D-1BE1-F9BC1704285A

treatment provided by

Plazi

scientific name

Necrophila (Calosilpha)
status

 

1) Necrophila (Calosilpha) spp.

The relative warps (RWs) of both male and female elytral apex of the four Necrophila (Calosilpha) taxa were calculated and plotted on an axis system (not shown here). The first RW axis represented 72.58% of shape variability and the second axis was 11.51%, thus indicating a significantly high sexual dimorphism. Due to these results, male and female taxa were further tested independently. Among the male groups, the first RW axis accounted for 49.69% of the total shape difference between the taxa, whereas the second axis accounted for 29.85%. Meanwhile, 58.98% of variability between the females was explained by the first RW axis and 14.11% by the second axis.

The scatter plot of the two first RWs for male Calosilpha displayed a greater overlap between the groups of the various taxa than in the females. The thin-plate spline (TPS) transformation grids indicated shape changes of the elytral apex on the first two RW axes, and in males it appeared more rectangular in all groups than in females (apart from in N. (C.) ioptera ). N. (C.) brunnicollis was slightly more truncate. In females, the shape of N. (C.) brunnicollis elytral apex was more robust. In N. (C.) cyaniventris and N. (C.) cyaneocephala , it was more prolonged and even pear-shaped ( N. (C.) cyaneocephala ). N. (C.) ioptera appeared subrectangular, more rounded and with more prominent tip than in the males ( Figs. 65–66 View FIGURE 65 View FIGURE 66 ).

Shape variety of the elytra between the studied taxa was indicated, as MANOVA for male groups revealed significant shape differences (F = 13.24; Wilk’s lambda = 0.1599; d.f. = 54/841; p <0.0001). Female groups demonstrated greater elytral shape variability (F = 19.56; Wilk’s lambda = 0.1001; d.f. = 54/907; p <0.0001).

To obtain separation of the studied groups, two individual canonical variate analyses (CVA) were performed, for male and female elytron apex, on the first 18 axes of the RW scores matrix that covered 99.76% of the shape variation between the males and 99.79% between the females. In both cases, the canonical axes indicated some separation of the taxa—especially of N. (C.) ioptera in both sexes—but generally there was an overlap between the groups ( Fig. 67 View FIGURE 67 ).

The jackknifed values of the confusion matrix ( Table 2) indicated that most (134 from 171) specimens of N. (C.) brunnicollis were assigned correctly; 8 had mean closer to N. (C.) cyaniventris , 27 to N. (C.) cyaneocephala , and only 2 specimens to N. (C.) ioptera . The majority of both N. (C.) cyaniventris and N. (C.) ioptera were assigned correctly. N. (C.) cyaniventris had 8 specimens closer in mean to N. (C.) brunnicollis and 3 specimens to N. (C.) ioptera of a total 39 specimens. N. (C.) ioptera had only 2 taxa closer to N. (C.) cyaniventris of a total of 37. Also, 33 N. (C.) cyaneocephala were assigned correctly of 56 while 14 were closer to N. (C.) brunnicollis , 7 to N. (C.) cyaniventris , and 2 to N. (C.) ioptera . This indicated N. (C.) brunnicollis and N. (C.) cyaneocephala are closer in their mean shape than are any of the other groups.

Females likewise showed an overlap between the groups. N. (C.) brunnicollis had a correct assignment in 148 specimens of 196 while 26 taxa had a mean closer to N. (C.) cyaniventris , 17 to N. (C.) cyaneocephala , and 5 to N. (C.) ioptera . While N. (C.) cyaniventris had 32 out of 44 correct assignments, 8 specimens were closer to N. (C.) brunnicollis and 4 to N. (C.) cyaneocephala . Further, 6 N. (C.) cyaneocephala were closer to N. (C.) brunnicollis and only 2 to N. (C.) cyaniventris , thus making 40 of 48 correctly assigned. Only 2 specimens of N. (C.) ioptera had mean closer to N. (C.) cyaniventris while 35 of 37 were correctly assigned. In general, N. (C.) cyaniventris and N. (C.) brunnicollis seem to be closer in elytral shape. The high misclassification in both sexes could be caused by the much larger number of N. (C.) brunnicollis samples used. N. (C.) ioptera demonstrated great separation from all other taxa.

Taxa Males Females

Correctly classified Misclassified Total Correctly classified Misclassified Total N. (C.) brunnicollis 134 37 171 148 48 196 N. (C.) cyaniventris 28 11 39 32 12 44 N . (C.) cyaneocephala 33 23 56 40 8 48 N . (C.) ioptera 35 2 37 35 2 37 When testing the effect size had on body shape, the natural logarithm of centroid size was used as an independent variable in the regression analysis. The results indicated significance in males (F = 21.621, p <0.0001), where size explained 22.47% of the shape dissimilarity. The permutation tests for 1,000 random permutations gave p <0.0259. In females, size explained a much higher percentage of 40.36% (F = 54.302, p <0.0001) and the permutation tests for 1,000 random permutations gave p <0.0147. Accordingly, it can be concluded that size has an irrefutable effect on shape variability among the studied taxa. Even though size explained a great deal of the difference between the groups, shape variability among them cannot be denied.

Kingdom

Animalia

Phylum

Arthropoda

Class

Insecta

Order

Coleoptera

Family

Silphidae

GBIF Dataset (for parent article) Darwin Core Archive (for parent article) View in SIBiLS Plain XML RDF