Moulting was initiated by the end of post-monsoon (January), continued up to summer (February-May) and until the onset of monsoon (June). Prior to the initiation of moulting, the weight of individuals increased, caused by high feeding activity and accumulation of excess water to develop pressure in the haemolymph to smoothen the exoskeleton. Based on the present observation, the moulting behaviour of an active pill-millipede (Figure 01a) can be classified into five stages: burrowing (Figure 01 b), construction of the moulting chamber (Figure 01 c), sedentary state (Figure 01 d), ecdysis (Figure 01 e) and regaining activity (Figure 01 f). Among the pill-millipedes studied, the smallest species A. disticta showed the highest percentage of moulting (51%) compared to the large-size species such as A. fumosa (41%), A. magna (21%) and A. dalyi (13%) (Figure 02). The duration required for initiation and termination of moulting in pill-millipedes was 4-5 days, but the moulted pill-millipede required more duration for regaining their original colour. For instance, A. disticta required a short duration (3-4 weeks), whereas other species required long duration (8-9 weeks) to regain the original colour.
Burrowing and moulting chamber: Usually, pill-millipedes feed on the shreds of leaf litter available on the upper strata of soil as long as the soil is moist. The mouthparts of pill-millipedes are suitable for burrowing in to the soil to make a protective shelter for moulting, almost similar to hibernation. They excavate the soil to make a tunnel and a moulting chamber will be constructed at the end of the tunnel (Figure 01d). In most instances, the construction of tunnels takes place at an angle of 50-60°. The depth of tunnel varies from 5-25 cm in the mesocosms and the diameter of moulting chamber usually fits to its partially-conglobated posture. The small pill-millipede A. disticta constructed narrow tunnels in shallow depth (about 5-10 cm), while large-size species burrowed deep (up to 15-25 cm). The maximum tunnel depth (faecal pellets + soil) attained in the mesocosms in our study was 25 cm and it is expected that the depth may increase in the natural habitats depending on the nature of soil and availability of moisture.
The moulting chamber is constructed by mixing faecal pellets, fragmented leaf litter and soil. Pill-millipedes preferred middle region of the mesocosm with moderate moisture for moulting chamber construction than the low-lying areas with more moisture. The inner wall of the chamber was smooth and highly polished (Figure 03a) unlike the hibernating chamber (quiescent state), probably due to continuous rubbing of the surface to eliminate exuviae.
Sedentary state and ecdysis: Prior to moulting, feeding activity of pill-millipedes will be the highest to meet the energy requirement and to accumulate sufficient amount of water for smoothing the exoskeleton. Once they enter the moulting chamber, its activity will be highly restricted and it will attain an almost sedentary state. During this stage, it prefers to be in conglobated posture (Figure 01d, Figure 03b-d). On initiation of moulting in mesocosms, their behaviour was observed by clearing the upper region (leaf litter, soil and faecal pellets). Sometimes, the moulting chambers were constructed at the edge of the mesocosm and the entire process of moulting was visible through the glass. On initiation of moulting, pill-millipedes attain partially conglobated posture (Figure 04 g, k). During ecdysis, exoskeleton shedding initiates from the head region and ends at pygidium or the anal shield. Moulting in all pill-millipedes studied was initiated with breakage of exoskeleton between collum and first pair of legs. The opened up collum with head portion (mouth parts, antennae, eyes and gnathochilarium) separates and projects upwards like a cap. The second breakage of exoskeleton initiates between the first pair of legs and extends towards the telopods. Subsequently, the pill-millipede attains fully conglobated posture and the exuviae shows partial attachment to the body. The exuviae stay, partially covering the back of pill-millipede for a few days, later it become brittle and breaks into pieces. At the end of ecdysis, either the whole exuviae or its pieces will be left in the moulting chamber as the tunnel exactly fits into its diameter (Figure 01 f). In A. dalyi and A. magna, the exuviae were rigid and stayed on the back (Figure 04 e, h), while in A. disticta and A. fumosa it quickly collapsed in to pieces (Figure 04 f, g).
Callow and regaining activity: After moulting, pill-millipedes become soft and light-coloured, with partially distinct bands on tergites called the ‘callow’ (Figure 04, e-l). At this stage, it grows and expands its size by using stored nutrients. During callow stage, pill-millipedes are easily accessible to predators (especially ants), thus they reside in the moulting chamber until the body sufficiently hardens (Figure 03b-d). At times, they come out soon after moulting, probably in search of a food source. The initial colour after ecdysisis was either bluish-green or sky-blue (A. disticta: Figure 04f, J; A. fumosa: Figure 04g, k), greenish tinge with light yellow bands (A. dalyi) (Figure 04e, i) and sky-blue with yellow tint (A. magna) (Figure 04h, i). After shedding the exuviae (within 2-3 days), the exoskeleton regains its strength and the normal colour will eventually be restored. Because of such colour variations in the late moulting stage, many pill-millipedes will be mistaken as new species or variants especially during the early monsoon season in the Western Ghats (Figure 04: compare adults a, b, c, d with respective post-callow stage m, n, o, p).
Figure 02. Percentage of moulting in four species of pill-millipedes in mesocosms (n=3; mean ± SD).
Figure 03. Moulting chamber and moulting pattern in Arthrosphaera magna in mesocosms: a. nature of moulting chamber; b. millipede housed inside the moulting chamber; c. ecdysis inside the moulting chamber (crumpled exuviae); d. final stage of moulting with exuviae on conglobated millipede.
Figure 04. Moulting stages of pill-millipedes: a, b, c, d. Arthrosphaera dalyi, A. disticta, A. fumosa and A. magna prior to moulting, respectively; e, f, g, h. early stages of moulting in respective species with major difference in the colour complex; i, j, k, l. moulted callow of respective species; m, n, o, p. conglobated stage of moulted individuals prior to leave the moulting chamber for normal activity. Note the colour difference between individuals before moulting (first column), during moulting (second and third columns) and after moulting (last column).
Moulting is common in arthropods and it is one of the essential requirements for growth, metamorphosis and population expansion (Ewer, 2005). Arthropods exhibit significant anatomical modifications to meet moulting requirements (Aguinaldo et al., 1997). During ecdysis, uptake of a large amount of water creates the turgid pressure on the exoskeleton for its rupturing (Kuballa and Elizur, 2007). Further, uptake of water facilitates expansion of a new exoskeleton leading to muscle regeneration. In our study, pill-millipedes ingested a large quantity of moist leaf litter to meet the demands of energy and water. Moulting strategies of pill-millipedes are interesting because of their low vagility and high-speed burrowing capacity (Ambarish and Sridhar, 2013). Most millipedes cope with the conditions of wet and dry soil regimes with physiological and behavioural strategies (e.g. active on the soil horizon or hibernation) (Dangerfield, 1998). Arthropods are also known to consume their own exuviae to restore minerals and energy (Ruppert et al., 2004). In one of the high altitude forests of the Western Ghats, an unidentified Arthrosphaera species ingested its own exuviae after moulting probably to replenish calcium and magnesium (Ashwini, 2003). None of the four pill-millipedes in the present study showed such behaviour in mesocosms.
It is known that moulting results in nearly 80% mortality in arthropods because of the callow stage or partial trapping in its own exuviae (Ruppert et al., 2004). Construction of moulting chambers by millipedes prevents easy accessibility for predators. In the soil strata of forests and plantations, sturdy moulting chambers are hidden and provide maximum protection for millipedes from their predators (e.g. ants, crow, wild boar and mongoose) (Eisner and David, 1967). The depth of moulting chambers in natural habitats may be deeper compared to mesocosms due to presence of predators. In many locations especially in Areca mixed plantations, the depth of the moulting tunnels varies depending on the species of pill-millipede (e.g. A. fumosa: 10-65 cm; A. magna: 30-60 cm). The depth of tunnels likely depends on the soil texture as well as the quantity of deposited organic matter.
Moulting is a complex phenomenon regulated by a series of hormonal changes and environmental cues (Loeb, 1993). Ecdysteroids are the moulting hormones responsible for ecdysis in arthropods (Spaziani et al., 1989).During moulting, cuticle proteins will be upregulated indicating that these proteins are essential for development and hardening of the exoskeleton (Wynn and Shafer, 2005). Cuticle proteins bind to the chitin so as to stimulate calcification (Kragh et al., 1997; Wynn and Shafer, 2005). The cryptocyanine transports many essential compounds like hormones, phenol and proteins to the hypodermis of arthropods during pre-moult stage (Terwilliger, 1999). Metallothionein is another important protein exists in haemolymph, which is necessary during pre-moulting and moulting stages (Syring et al., 2000). According to Ewer and Reynolds (2002), arthropods generally increase their neuropeptide (20E) concentration to initiate the moulting process by stimulating epidermis to divide, separate from the old cuticle and synthesize/secrete the new cuticle. There is a decrease in 20E concentration in arthropods leading to completion of moulting process. 20E is known to act on the nervous system to initiate moulting and also influences peripheral tissues to harden the newly synthesized cuticle. Such proteins and hormones are yet to be characterized in pill-millipedes of the genus Arthrosphaera as well as the other genera of Sphaerotheriida. However, development of dorso-ventral segmentation pattern in temperate pill-millipede belonging to the genus Glomeris have been investigated recently by Janssen et al. (2004).
This is the first report on moulting behaviour of pill-millipedes of the genus Arthrosphaera belong to the order Sphaerotheriida. Among the four pill-millipedes studied, the extent of moulting ranged between 13-51% during the summer season in the mesocosms. After moulting, the pill-millipedes exhibited different colours with less prominent striations leading to mistaken identity as new species or variants especially during early monsoon in the Western Ghats. In addition, to understand the physiological aspects of moulting, studies on the cascade of unclear events responsible for the regulation of ecdysis will be more rewarding. Molecular studies aiming at the hormonal regulation of moulting in pill-millipedes especially microarray techniques may develop a method for controlling the moulting cycle. Understanding such control mechanisms improves knowledge of the life cycles of pill-millipedes and will help in large-scale cultivation and production of organic manure under in situ and ex situ conditions.
The authors are grateful to Mangalore University for permission to carry out this study in the Department of Biosciences. CNA acknowledges the award of Inspire Fellowship by the Department of Science and Technology, New Delhi, Government of India (DST/INSPIRE Fellowship/2011/294: Award # IF110540). KRS acknowledges the award of UGC-BSR Faculty Fellowship by the University Grants Commission, New Delhi, India. The authors are thankful to the editor and the anonymous reviewers for genuine suggestions to improve the earlier version of the manuscript.
Aguinaldo A.M.A., Turbeville J.M., Linford L.S., Rivera M.C. and Garey J.R. (1997) Evidence for a clade of nematodes, arthropods and other moulting animals. Nature 387: 489–493.
Ambarish C.N. and Sridhar K.R. (2013) Observations on pill-millipedes of the Western Ghats (India). Journal of Agricultural Technology 9: 61–79.
Ashwini K.M. (2003) Ecological Studies on Indian Pill Millipede, Arthrosphaera magna. Ph.D. Dissertation, Mangalore University, Mangalore, India.
Ashwini K.M. and Sridhar K.R. (2005) Leaf litter preference and conversion by a saprophagous tropical pill millipede, Arthrosphaera magna Attems. Pedobiologia 49: 307-316.
Attems C. (1936) Diplopods of India. Memoirs of Indian Museum 11: 133–167.
Dangerfield J.M. (1998) Biology and ecology of millipedes in the Kalahari. Transactions of the Royal Society of South Africa 53: 183–194.
Eisner T. and David J.A. (1967) Mongoose throwing and smashing millipedes. Science 155: 577–579.
Ewer J. (2005) How the Ecdysozoan changed its coat. PLoS Biology 3: 1696–1699.
Ewer J. and Reynolds S. (2002) Neuropeptide control of molting in insects. In: Pfaff D.W., Arnold A.P., Fahrbach S.E., Etgen A.M. and Rubin R.T. (eds). Hormones, Brain and Behavior. Academic Press, San Diego, 1–92 pp.
Golovatch S.I. and Kime R.D. (2009) Millipede (Diplopoda) distributions: A review. Soil Organisms 81: 565–597.
Hopkin S.P. and Read H.J. (1992) The Biology of Millipedes. Oxford: Oxford University Press.
Janssen R., Prpic N.M. and Damen W.G.M. (2004) Gene expression suggests decoupled dorsal and ventral segmentation in the millipede Glomeris marginata (Myriapoda: Diplopoda). Developmental Biology 268: 89–104.
Jurenka R. (2007) Insect physiology. McGras-Hill Encyclopedia of Science & Technology 9 (10th Edition). Parker S.P. (ed). McGraw-Hill Book Company, New York. 323–353 pp.
Kadamannaya B.S. and Sridhar K.R. (2009) Morphology and distribution of four pill millipedes (Arthrosphaera) of the Western Ghats. Animal Biology Journal 1: 81–98.
Kadamannaya B.S., Ambarish C.N. and Sridhar K.R. (2012) Morphological features of three endemic pill-millipedes of the genus Arthrosphaera in Western Ghats of India. Animal Biology Journal 3: 181–193.
Kragh M., Molbak L. and Anderson S.O. (1997) Cuticular protein from the lobster, Homarus americanus. Comparative Biochemistry and Physiology B 118: 147–154.
Krause D.W. (2003) Late cretaceous vertebrates of Madagascar: A window into Gondwanan biogeography at the end of the age of dinosaurs. In: Benstead S.M. and Goodman J.P. (eds), The Natural History of Madagascar. Chicago University Press, Chicago, 40–47 pp.
Kuballa Z. and Elizur A. (2007) Novel molecular approach to study moulting crustaceans. Bulletin of Fishery Research Agency 20: 53–57.
Loeb M.J. (1993). Hormonal control of growth and reproduction in the arthropods. American Zoologist 33: 303–307.
Pocock R.I. (1899) A monograph of the pill-millipedes (Zephroniidae) inhabiting India, Cylon and Burma. Journal of Bombay Natural History Society 12: 269–285.
Ruppert E.E., Fox R.S. and Barnes R.D. (2004) Invertebrate Zoology (7th Edition), Brooks/Cole, USA.
Shelley R.M. (2007)Taxonomy of extant Diplopoda (millipedes) in the modern era: Perspectives for future advancements and observations on the global diplopod community (Arthropoda: Diplopoda). Zootaxa 1668: 343–362.
Sierwald P. and Bond J.E. (2007) Current status of the Myriapod Class Diplopoda (millipedes): Taxonomic diversity and phylogeny. Annual Review of Entomology 52: 401–420.
Spaziani E., Watson R.D., Mattson M.P. and Chen Z.F. (1989) Ecdysteroid biosynthesis in the crustacean Y-organ and controlled by an eyestalk neuropeptide. Journal of Experimental Zoology 252: 271–282.
Sridhar K.R. and Ambarish C.N. (2013) Pill-millipede compost: A viable alternative toutilize urban organic solid wastes. Current Science 104: 1543–1547.
Sridhar K.R., Kadamannaya B.S. and Karamchand K.S. (2013) Nutrient composition of pill millipedes manure of Western Ghats, India. Journal of Forestry Research 24: 539–545.
Syring R.A., Hoexum B.T. and Brouwer M. (2000) Cloning and sequence of cDNA encoding for novel copper specific Metallothionein and two cadmium inducible Metallothionein from the blue crab, Calinecteus sapidus. Comparative Biochemistry and Physiology C 125: 325–332.
Terwilliger N.B. (1999) Haemolymph protein and molting in crustaceans and insects. American Zoologist 99: 589–599.
Wesener T. and Sierwald P. (2005) The giant pill-millipede of Madagascar: Revision of the genus Spheromimus, with a review of morphological terminology (Diplopoda, Sphaerotheriida, Sphaerotheriidae). Proceedings of the California Academy of Sciences 56: 557-599.
Wesener T. and van den Spiegel D. (2009) A first phylogenetic analysis of giant pill-millipedes (Diplopoda: Sphaerotheriida), a new model Gondwanan taxon, with special emphasis on island gigantism. Cladistics 25: 1–29.
Wesener T., Raupach M.J. and Sierwald P. (2010) The origins of the giant pill-millipedes from Madagascar (Diplopoda: Sphaerotheriida: Arthrosphaeridae). Molecular Phylogeny and Evolution 57: 1184–1193.
Wynn A. and Shafer T.H. (2005) Four differentially expressed cDNA in Calinecteus sapidus containing the Rebers-Riddiford consensus sequence. Comparative Biochemistry and Physiology B 141: 294–306.