The previous study showed that the red coloration of the ventral skin of the Japanese newt Cynops pyrrhogaster was associated with the number of carotenoid vesicles and the content of carotenoid in the pigment cell of the skin. To elucidate the mechanism for the red coloration of the skin of the newt, we studied the food habit of the juvenile from the Japanese newt Cynops pyrrhogaster. Sixty-two juveniles were collected in Fukue Island in Nagasaki Prefecture from November 2000 to May 2002 and divided into 2 groups according to the snout-vent length (SVL). Over 400 prey animals were obtained from the juveniles by stomach flushing. In the larger group (SVL>30.0mm), Collembola (45.4%) and Acari (12.6%), which are very common species of soil animals, were the prey animals dominant in number. In the group with the smaller SVL (<29.9mm), Collembola (30.4%) and Acari (25.4%) were in number as well. We also studied the food habit of the Japanese clouded salamander, Hynobius nebulosus. In the salamander, Doratodesmidae (56.5%) and Amphipoda (13%) were the prey animals dominant in number. Our results, taken together, suggest that the Japanese juvenile C. pyrrhogaster does not change its food habit as it grows, and that it eats soil animals common in its habitat. Moreover, the food habit of juvenile C. pyrrhogaster differs from that of H. nebulosus, although the juveniles of both species live in the same area.
INTRODUCTION
The color of the ventral skin of the Japanese newt C. pyrrhogaster is creamy at metamorphosis, but it turns red as the animal grows. In amphibians, 3 kinds of pigment cells, i.e., melanophores, xanthophores and iridophores, can be seen in the ventral skin of the juvenile after metamorphosis (Bagnara, 1998). We have shown that adult newts have many carotenoid vesicles in the xanthophores in the red skin (Matsui et al., 2002), and that the number of carotenoid vesicles in the pigment cell was very low in the ventral creamy skin at metamorphosis, but increased in the red skin (Matsui et al., 2003). The carotenoid content was also very low in the creamy skin of small juveniles, but dramatically high in the red skin of adult newts (Matsui et al., 2003). Based on these findings, it was concluded that the increase in the size and the number of carotenoid granules and also the carotenoid content in the ventral skin were very important for red body coloration during the growth of this species (Matsui et al., 2003).
There are two ways for formation of the body coloration in animals. One is that body color is produced by auto-synthesis of pigments. The other takes place by the uptake of pigments directly from foods. Since animals are unable to synthesize carotenoids by themselves (Goodwin, 1986), they must take up pigments via foods for their body coloration. To understand the reason for this change in coloration of the ventral skin of the newt, it is necessary to investigate the food habit of this species. Here we report for the first time the food habits of young juvenile C. pyrrhogaster having the ventral creamy skin and other ones with the red skin. The food habit of H. nebulosus is also reported.
MATERIALS AND METHODS
Animals
Juveniles of the Japanese newt C. pyrrhogaster were collected from November 2000 to May 2002 on the dike that was slightly elevated trail between adjoining paddy fields in Fukue Island in Nagasaki Prefecture, Japan. The juveniles were divided into 2 groups based on their snout-vent length (SVL; SVL>30.0mm and SVL<29.9mm). Prey animals were obtained separately from 62 individuals by the stomach flushing and fixed in 70% ethanol (Joly and Giacoma, 1992; Ihara, 1998). Since stomach items were enveloped in an opaque membrane, the membrane was removed by using fine-tipped forceps to identify the ingested species under a stereoscopic microscope (Aoki, 1999). The sample was then put on a piece of filter paper (Whatman No. 5) to eliminate an excess of 70% ethanol around the body. Total weight of prey animals in the same family was determined first using an electronic balance (Sartorius, model BP150), since each animal was too light to be weighed. Then, the average weight of each prey animal was calculated by dividing the total weight by the number of prey animals weighed. After the stomach flushing, the juveniles were released at the same location at which they had been captured.
Sampling of soil animals
Soil animals were obtained in May 2002 from the habitat of the juvenile newts and salamanders. The soil was scooped from 2 grids (W×L×D=25×25×5 cm). Then, animals were collected routinely by exposure of the soil to a 40-watt bulb for 48 hr using the Tullgren funnel.
Determination of the body color of newts
The body color of 21 juveniles collected in May 2002 was determined under daylight using the Color chart 1368 printed in the CMYK style (Optic design laboratory, Tokyo) (Sawada, 1963).
Statistical analysis
Kendall's rank correlation coefficient (Sokal and Rohlf, 1973) was used to test the significance of the food palatability and the chi-square (χ2) test (Sokal and Rohlf, 1973) for the significance of the food compositions between the smaller and larger juveniles.
RESULT
We identified 402 prey items washed out from the stomach of 62 juvenile newts that were collected on Fukue Island in Nagasaki Prefecture. As shown in Fig. 1, prey animals in the stomach were trapped within an opaque envelope. However, there was no difficulty to identify them under a stereoscopic microscope after the envelope had been removed. Table 1 shows the number of newts and salamanders, which we collected for this study, and the SVL and the mouth width of them. Fifty-eight out of the 62 newts had the stomach contents (Table 1). Table 2 shows the relationship between the size (SVL) and the color of the ventral skin of the juvenile newts. The color of the ventral skin of the smaller juveniles (S) appeared creamy, but 9 out of 10 larger juveniles (L) had the ventral red skin. Table 3 shows food habits of the newt C. pyrrhogaster and the salamander H. nebulosus. In the smaller juveniles (SVL<29.9mm), Collembola and Acari, common in the soil, were the prey animals dominant in number, representing 30.4% and 25.4%, respectively of the ingested prey (Table 3). Statistical analysis of the animals extracted from the stomach and those living in the soil gave a Kendall's rank correlation coefficient of 0.451 (p=0.019), indicating a good correlation between the two. However, Collembola constituted only 7.9% and Acari 7.7% of the total weight of the stomach content extracted from the smaller juvenile newts (Table 3). In the larger juvenile newts (SVL>30.0mm) as well, Collembola (45.4%) and Acari (12.6%) were dominant, and the former constituted only 8.5%, and the latter, 2.7%, of the total weight of the stomach content as well (Table 3). No significant difference was observed in the diet compositions between the smaller and larger juveniles (the χ2-test; χ2=37.07, df=12, p=0.065). Furthermore, we could not find any seasonal changes in the food palatability either. The newts ate Collembola and Acari mainly in number in spring and autumn. As also seen in Table 3, Doratodesmidae (56.5%) was the most dominant prey animal of H. nebulosus in terms of number, but constituted only 0.9% of the total weight of the stomach content (Table 1). On the other hand, Megascolecidae (4.4%) was numerically a minor prey animal of H. nebulosus, but constituted 90.5% of the total weight of the stomach content. Neither Doratodesmidae nor Megascolecidae was found among prey items extracted from the stomach of C. pyrrhogaster.
Fig. 1
Prey animals extracted from the stomach of a juvenile newt. Prey animals prior to (1) and after (2) removal of the envelope are shown in this figure. Each prey animal in the stomach was enveloped in the light brown membrane and the prey had coalesced into a mass. a, Araneae. b, Collembola. c, Acari. Bar=1mm.
Table 1.
Number and size of C. pyrrhogaster and H. nebulosus collected.
Table 2.
Relationship between the size and the color of the ventral skin of juvenile newts.
Table 3.
Dietary compositions of C. pyrrhogaster and H. nebulosus collected from the same area in the paddy field in Fukue Island.
DISCUSSION
In the present study, diet compositions of Japanese newts during the juvenile period were examined. The juvenile newt C. pyrrhogaster consumed Acari and Collembola as their main prey animals, whereas the salamander H. nebulosus preferred Doratodesmidae. Doratodesmidae was not observed in prey items extracted from the stomach of the juvenile newts. To date there has been only 1 report showing the food palatability of the juvenile and adult salamander Hynobius tokyoensis, a Japanese urodeles (Ihara, 1998). According to Ihara (1998), this salamander eats Porcelionidae and Trochelipidae. To the best of our knowledge, this is the first report describing the food habits of juvenile C. pyrrhogaster and H. nebulosus. Food habits of several species in the foreign urodela have been reported (Hamilton, 1932; Morgan and Grierson, 1932; Martof and Scott, 1957; Bury and Martin, 1973; Powders and Tietjen, 1974; Burton, 1976; Verrell, 1987; Wisniewski, 1989). Adults of 5 foreign species of urodeles have food palatability different from one other (Burton, 1976). Regarding the food habits of foreign juvenile salamanders, there is 1 report showing that the size of prey animals becomes bigger as juvenile salamanders grow, but that the juveniles do not change the food palatability (Maglia, 1996). In our study the smaller and larger juvenile newts ate Acari and Collembola dominantly as their prey animals. Thus, it is reasonable to conclude that C. pyrrhogaster does not change its food habit during the juvenile period. Since Acari and Collembola are very tiny and light, however, they constitute only a few percents of the total weight of the stomach content extracted from C. pyrrhogaster. Therefore, to keep active, juvenile newts need to eat other heavier soil animals such as Coleoptela, Gastropoda, Aphidoidae, Amphipoda and Diptela. In this regard, H. nebulosus eat Megascolecidae. Thus, the food habit of juvenile C. pyrrhogaster clearly differs from that of H. nebulosus, although juveniles of the 2 species inhabit in the same area.
As mentioned earlier, the color of the ventral skin of smaller juveniles is creamy, but that of larger juveniles is red (Matsui et al., 2003). The color of the ventral skin of the Japanese salamander H. nebulosus is different from that of the newt C. pyrrhogaster. To change the color of the ventral skin red, juvenile newts need to eat foods containing carotenoids, because they cannot synthesize carotenoids by themselves (Goodwin, 1986). The red coloration of the ventral skin of C. pyrrhogaster was closely related to the size and the number of carotenoid granules and the content of carotenoids in the granules (Matsui et al., 2003). As the juvenile newts do not change the food palatability during the red body coloration, the body coloration may be caused by the constant accumulation of carotenoids in the skin. However, we cannot exclude the possibility that older juvenile newts eat soil animals carrying a large amount of carotenoids, but they may have not been found in this study. To prove this, we must investigate which prey animals contain larger amounts of carotenoids, and what kinds of carotenoids are contained in the prey animals. Nevertheless, it is evident that the food habit of C. pyrrhogaster differs from that of H. nebulosus. The red body coloration of the newt C. pyrrhogaster is probably not due to a change in the food habit of the animal during the juvenile period.
