Keywords
body size, variation, discontinuity, phenotypic plasticity, dimorphism
Introduction
Luciola parvula(Coleoptera, Lampyridae) is a terrestrial firefly species widely distributed in Japan and shows large geographical variation in adult body size especially in mountainous areas. Previous studies reported that there were two ecological morphs of adults of this species: large and small morphs. However, the procedure used to identify the two morphs has remained ambiguous and confused. Therefore, the present article begins with a brief review of the two morphs in this species and then proposes an allometric approach to detect the existence of the two morphs.
As far as the present author knows, Ohba (1) was the first to describe the criteria for the classification of the two morphs in this species. He collected male adults at Mt. Hakone, central Japan and then identified the large morph above 700 m altitude and the small morph below this altitude. He also reported that the approximate body length was 10 mm for the large morph and 6 mm for the small morph.
Ohba (2) further examined male adults of this species throughout Japan and then found the large morph with body length 7 to 9 mm and pronotum width larger than 2.1 mm and the small morph with body length 5 to 7 mm and pronotum width smaller than 2.1 mm.
Mitsuishi (3–6) collected male adults of this species at 41 sites in Nagano Prefecture, central Japan and classified them into the large morph (body length larger than 8 mm) and the small morph (body length approximately 7 mm). Aa a result, he found that this species in this prefecture tends to be segregated into a northeastern and a southwestern group corresponding to the large and the small morph, respectively, but that both morphs exist above 1000 m altitude. However, these findings were not consistent with Ohba’s findings (1). This may be partly due to the arbitrariness of the criteria for the two morphs.
Ohba (7) reviewed his studies on the two morphs of this species and summarized his criteria for the classification of them: the large morph with body length 7 to 8 mm below 1600 m altitude and the small morph with body length 5 to 6 mm below 800 m altitude.
Kusaoke et al. (8) investigated the distribution of this species in Toyama Prefecture, adjacent to Nagano Prefecture mentioned above and found the small morph (body length approximately 6 mm) at three sites at altitudes of 400 to 700 m and the large morph (body length approximately 8 mm) at an altitudes of 1100 m. As they mentioned, these findings almost fell into Ohba’s criteria (7). Therefore, they considered that Mitsuishi (3–6) had found only the large morph above 1000 m altitude in Nagano Prefecture.
However, Kusaoke et al. (8) also found at Mt. Daisen, Tottori Prefecture, that the large morph (mean body length 8.12 mm) exited at 1709 m altitude and small morph (mean body length 7.14 mm) at 800 m altitude. These mean values of body length were not consistent with Ohba’s criteria (7).
In these studies, this species was more or less arbitrarily classified into the large and small morphs. No statistical analysis was performed to examine whether or not body size varies continuously.
Molecular phylogenetic studies suggested that the small morph originated from an ancestor similar to the large morph (9, 10). However, no genetic difference was found between the two morphs at Mt. Daisen, Tottori Prefecture (8).
Ohba (7) also showed that the flash interval (0.5 to 0.6 s) of the small morphs is smaller than that (0.7 to 0.8 s) of the large morph. However, he did not take into consideration the effect of temperature on flash intervals. In fact, such temperature effects were found and statistically analyzed in Nipponoluciola cruciate (formerlyLuciola cruciate ) (11–13). Therefore, it remains unclear whether the two morphs of this species show different flash intervals under controlled temperature.
Regarding morphological measurements of L . parvula , the previous studies focused on simple descriptive statistics such as mean values of body length. However, multivariate analyses are considered necessary for detecting size dimorphism in this species. Therefore, the existence of inconsistent criteria for the two morphs of this species leads the present author to reexamine published data on measurements of this species.
Fortunately, Mitsuishi (3–6) showed his measurement data on body length, body width and pronotum width at 41 sites in Nagano Prefecture. Using these data, the present article attempts to detect differences in scaling relationships between the two morphs by allometric analysis. The results will help to provide direction for future studies on the size dimorphism of this species.