Abstract: The Longgupo mandible was identified as belonging to Homo erectus by Huang et al. ( 1991) and w as suggested to be close to Homo ergaster by Huang et al ( 1995) . The hominid status has been challeng d by Wolpoff ( 1996) , Schwartz and Tattersall ( 1996) , Pope ( 1998) as well as Etler and Zhou ( 1998) who indicated an ancestor /descendant relationship between Lufengpithecus and the Longgupo mandible. But there is no detailed discussion in these literatures. The present author would like to present a series of comparisons between the Longgupo mandible and various East African early Homo as well as the mandible from Dmanisi to discuss the attribution of the Longgupo mandible. Table 1 shows that the width and height of Longgupo mandible is much smaller than those of East African early Pleistocene hominids including ER992 and W T15000 which are considered as belonging to Homo ergaster by some scholars including Huang et al. Huang et al ( 1995) indicated that the bucco-lingual expansion of the lower fourth premolar makes Longgupo mandible similar to the Homo ergaster ER992. Table 2 shows that the Longgupo premolar is much smaller than the East African early Pleistocene hominids, while the dimensions of Longgupo premolar is just within the variation range of that o f Lufengpithecus found in Yuanmou, Yuannan. ( some one identified it as Lufengpithecus yuanmouensis). That the width is long er than the length of the premolar crown is not only shown in Longgupo premolar and that of Lufengpithecus but it is also shown in Dryopithecus kaiyuanensis in which the ratios of width to length of the crown of fourth premolar in two individuals are 11. 5 and 10. 8. From Table 3 it is obvious that the dimension o f Longgupo molar is concordant very well with that of Lufengpithecus found in Yuanmou and quite smaller than that of all early Pleistocene hominids in East Africa. Huang et al have indicated that the Longgupo premolar has large talonid basin which occupies 2 /3 of the total area of the occlusal surface. On the basis of the figure printed in the article written by Huang et al ( 1991) the present author estimated the ratio of its talonid length to the length of the tooth being about 67% . Table4 shows that the ratio of talonid length to the leng th of the tooth in fourth premolar is around one third in all compared East African early Pleistocene hominid premolars, so the Longgupo specimen is quite different in this feature from the latter specimens. From the figure in page 164 of the article written by Brown and Walker ( 1993 ) it a ppears that the ratio of this kind in W T15000 is close to one third. On the other hand, in Lufengpithecus and Dryopithecus kaiyuanensis found in Yunnan, the ratio of this kind is about 1 /2 or even larger. It is more reasonable to consider the large talonid of Longgupo fourth premolar suggesting its affinity with apes rather than being a retension of ape feature in humams as Huang et al. ( 1996) suggested. Huang et al. ( 1995) mentioned that the crown of Longgupomolar is low but had not given the measurement. The present author estimated the crown height of both molar and premolar are 6. 2mm on the basis of the figure provided in their article. In comparison with the crown height of WT15000 ( left, 7. 6mm; right, 7. 7mm for molars; 9. 4mm and 9. 1mm for the fourth premolars o f both sides) and OH13 ( 9, 1mm for premolar) , the fourth premolar and first molar of Longgupo mandible are very low The ratio of crown height to crown length in Longgupo molar and premolar a re 0. 56 and 0. 84 respectively. These are quite lower than those in W T15000 which are 0. 61 for first molars of both sides and 1. 04 and 1. 01 for premolars of both sides respectively. The ratio in Longgupo is also lower than that in Homo habilis. Huang et al. ( 1996) mentioned in page 202 that the Longgupo teeth have thick enamel and emphasized that this is a derived character of hominid. In their article published in 1995 ( page 277) they mentioned the enamel of Longgupo molar is relatively thin. but they have not provided the measurement of the enamel thickness in both papers. Wu Rukang et al. ( 1985) had reported that the Lufengpithecus lufengensis has thick and very thick enamel. The enamel of Afropithecus trukanaensis is also very thick. Therefore whether the enamel is thick or not is of no help to attribute Longgupo specimen to hominid. Huang et al. ( 1995) indicated that the Longgupo premolar is close to the East Africa n early Pleistocene hominids because of the bifurcation of the root shown in both Longgupo and ER992. But the roots of fourth premolar of Lufengpithecus and Dryopithecus kaiyuanensis are also bifurcated, while it is not bifurcated in WT15000. As Wolpoff et al. ( personal communication) has pointed out that the anterior contact facet of the fourth premolar of Longgupo mandible is situated lingually rather than centrally on the mesial surface of the tooth. This is an important feature of ape rather than hominid Table 5 shows that the Longgupo dentition is much smaller than that of the Dmanisi mandible and the molars of these two specimens are quite different in shape as shown by the length-width index. Dmanisi mandible is 20mm thick at the level of first molar. It is 28. 5mm high a t the level of the mental foramen. Its mandibular height at the level of first molar can not be measured because of damage. But we could estimate that the mandibular height at M1 level could be lower than that at the level of mental foramen in reference to the cases in both East African early Pleistocene hominids and Lantian Homo erectus. So it is reasonable to suppose that the robustness index of Dmanisi mandible at M1 level is probably larger than 70. 2 ( 100 X 20 /28. 5) . This is quite different from that of Longgupo mandible which is 64. 2. O H16 is the East African early Pleistocene hominid which preserved both upper incisor and lower cheek teeth. In this specimen, the ratios of crown areas of first upper incisor to first lower molar and fourth lower premolar are 1 /1. 91 and 1 /1. 20. If Longgupo mandible belong ed to a creature similar or close to East African early Pleistocene hominid, the crown area of its first upper incisor could be supposed as close to 58. 7 square mm or 56. 3 square mm based on the calculation on the basis concerning to the molar and premolar respectively. But the crown areas of Homo erectus first upper incisors found at Yuanmou, Yunnan a re 92. 3 square mm ( left) and 98, 9 square mm ( rig ht). According to the published data available the Yuanmou incisor is later than Longgupo mandible by about 100-300 thousand years. If Longgupo mandible is the predecessor of Yuanmou Homo erectus, the incisor of Longgupo should enlarge extraordinarily rapidly as such in so short time. It seems that the process of enlargement as such is unlikely. So the Longgupo hominoid could not be ancestral to Homo erectus. As the author has analysed in the present paper, the features on which the conclusion of Huang et al. ( 1995) are based are composed of features shared by both hominids and apes, the features seldomly present in hominids but commonly existing in apes, and features unique for apes but no t present in early Pleistocene hominids. As to the dimensions and shape indices Longgupo mandible and teeth are much smaller and different from those of East African early Pleistocene hominids and Dmanisi mandible, but it is concordant with those of Lufengpithecus found at Yuamou, Yunnan. The difference between Longgupo and the African hominids is very difficult to be explained by normal variation. Although it is not sure whether there was an ancestor /descendant relationship between Lufengpithecus and Longgupo mandible because of the fragmentary state of it , but it is no doubt to attriute it within the circle of ape instead of hominid. In a recent paper Ciochon wrote: “ For some Western scientists the teeth ( premolar and molar of Longgupo hominoid mandible, no ted by the present author) share features with earliest Homo in East Africa—— leading us to suggest a direct link, a “ dispersal” of African hominins to East Asia about 2 million years ago. But Chinese paleoanthropologists tend to see these primitive features as deriving from Asian apes and suggest a local Asian origin for Homo erectus” ( Ciochon, 2000 ). But what happened in China is, paleoanthropolgists who see these primitive features as deriving from Asian apes, such as Zhou ( Etler and Zhou, 1998) and the present author ( Wu, 1999a b; W u et al. , 1999) do not suggest a local Asian origin for Homo erectus but suggest to expel the Longgupo mandible from the human circle and attribute the mandible to Asian ape. Indeed, there are a few Chinese literatures suggesting not to completely exclude the potential possibility of the Asian origin of hominid and encouraging people making efforts to search for reliable fossil evidence, but this does not imply to suggest the local origin of Homo erectus. In fact, only very few Chinese researchers misinterpreting the available fossil data advocate the local origin of Homo erectus.

Key words: Mandible, Ape, Longgupo