WINGS, SANDER, P.M., TUTKEN, T., FOWLER, D.W., & SUN, G. (2007), Growth and life history of Asia's largest dinosaur, Journal of Vertebrate Palaeontology 27(3, supp): 167A

ABSTRACT

A new extensive dinosaur locality was discovered during fieldwork of the Sino-German Project in April 2006. The site is located 24 km NNE of Jiangjunmiao ( Junggar Basin ), in the upper Shishugou Formation (Upper Jurassic). Despite exploration of this site being cancelled prematurely due to permit problems, we were able to excavate and study several important specimens including associated and partly articulated sauropod and theropod remains. Among these finds is an enormous mamenchisaurid (Sauropoda) ulna (preserved length: 96 cm), representing the largest dinosaur known from Asia . Using current methods for sauropod mass estimation, and comparison to published Mamenchisaurus data, we calculated a mass estimate of 30,868 kg (assuming a specific density of 0.8 g/cm 3 ). Sectioning of the ulna revealed a very thick primary cortex with very little bone remodeling and a small medullary cavity, allowing a detailed study of bone histology and isotope geochemistry. The primary cortex consists of laminar fibrolamellar bone, as in other sauropods. However, the specimen differs in the early onset of cyclical growth, at 40% final size. Growth cycles are delimited by lines of arrested growth (LAGs), and a total of 27 were counted. By calculating body mass increase from cortical thickness increase, we derived the first growth curve for a sauropod. Maximum mass gain occurs between cycles 13 and 14 and is 2006 kg. The last three LAGs are very closely spaced, indicating that the individual was fully grown. A cross-section of the cortex was serially sampled at mm-resolution for analysis of carbon and oxygen stable isotopes in bone apatite. d 13 C values (–7.3±0.4 VPDB, n=29) indicate that the mamenchisaurid fed on C3 plants. d 18 O CO3 values (–12.5±0.5 VPDB, n=29) reflect apatite formation from bodyfluids with low d 18 O values. The cortex displays cyclic intra-bone variation of the d 18 O values which is unlikely to result from diagenesis and thus probably reflects an in vivo signal. However, the number of d 18 O cycles (n=11) does not correspond to that of LAGs observed (n=27). Relation between annual growth marks and d 18 O seasonality will be further tested with a higher spatial resolution.