18 The Philippine Journal of Science 1923 



rupted by deep arms of the sea since the early Tertiary at 

 least. The "fairly definite western frontier" of Barbour 

 is Wallace's Line. In the unstable area land connections 

 between what are now individual islands have been inter- 

 mittent and, as between these islands and the lands to the east 

 and west, the connections have apparently never been more 

 than narrow isthmuses. In other words, intermigrations in the 

 entire region from Lombok and Celebes eastward to New Guinea 

 and northward through the Philippines have been inhibited by 

 the generally constant archipelagic condition of the entire region. 

 In reference to Borneo and Celebes the Sarasins 22 state that, as 

 Celebes and Borneo do not present a single animal in common 

 that is not found also in Java, Sumatra, or the Philippines, there 

 is not the slightest possibility that a direct land bridge ever 

 existed between Celebes and Borneo across the Macassar Strait 

 since early Tertiary times. That there were indirect connections 

 via the Sulu Archipelago, Mindanao, and the Sangi Islands to the 

 north, and between Celebes and eastern Java by way of the Pos- 

 tilion and Paternoster islets, Bali and Lombok to the south, is 

 entirely probable ; in fact, almost certain. 



I cannot accept Schuster's 23 general conclusions regarding the 

 climate and vegetation of Pleistocene times in Java, nor his 

 explanation of the geologic sequence of land connections in 

 eastern Malaysia. Making due allowance for some manifestly 

 erroneous identifications on his part, 24 I can see no reason for 

 considering that, at the time the Trinil beds were formed, the 

 low-altitude Javan climate was cooler than it is to-day. The 

 altitudinal range of many of the species listed by him does not 

 conform to their actual occurrence in nature; most of them, 

 both in Java and in the Philippines (so far as they occur here)] 

 are low-altitude forms, even though some may exceptionally 

 extend to and above an altitude of 1,200 meters. That in Pleis- 

 tocene times the forests of Java were "typische Regenwalder 

 der gemassigten Zone" positively cannot be accepted. His own 



22 Sarasin, P. und F., Materialen zur Naturgeschichte der Insel Celebes 

 III. Ueber die geologische Geschichte der Insel Celebes auf Grund der 

 Thierverbreitung (1901). 



23 Schuster, J., Monographie der fossilen Flora der Pithecantropus- 

 Schichten (1911) 1-70, t. 1-27. Reprint from Abhandl. Kgl. Bay Akad 

 Wissensch. Math.-Phys. Klasse 25 6 (1911). 



"Thus, Viburnum coriaceum. While perhaps the correctness of the 

 identification cannot be proved or disproved, the figure may just as well 

 represent some species in any one of a half-dozen other genera in as many 



