[163] These interesting fossils are figured and described in the Am. Journal of Conchology.
[164] "The whole basin between the Orinoco and the Amazon is composed of granite and gneiss, slightly covered with débris. There is a total absence of sedimentary rocks. The surface is often bare and destitute of soil, the undulations being only a few feet above or below a straight line."—Evan Hopkins, in Quart. Jour. Geol. Soc., vol. vi.
[165] See Darwin on the absence of extensive modern conchiferous deposits in South America, Geological Observations, pt. iii., ch. v.
[166] Agassiz calls the average temperature 84°, which, it seems to us, is too high. The mean between the temperatures of Pará, Manáos, and Tabatinga is 80.7°.
[167] Buttress roots are not peculiar to any one species, but common to most of the large trees in the crowded forest, where the lateral growth of the roots is made difficult by the multitude of rivals. The Paxiuba, or big-bellied palm, is a fine example.
[168] Von Martius, in his great work on the Brazilian Palms, enumerates in all 582 species.
[169] The Portuguese and Brazilians call it seringa, or syringe, in which form it is still used extensively, injections forming a great feature in the popular system of cures. The tree mentioned above yields most of the rubber of commerce, and is considered distinct from the species in Guiana, S. elastica; while the rubber from the Upper Amazon and Rio Negro comes from the S. lutea and S. brevifolia. Agassiz puts milk-weed in the same family!
[170] We await the Professor's examination of his "more than 80,000 specimens" before we give the number of new species.
[171] The natives have this notion about the land-tortoise, that by throwing it three times over the head, the liver (the best part) will be enlarged.
[172] Dephinus pallidus. Bates observed this species at Villa Nova; we saw it at Coary, 500 miles west; and Herndon found it in the Huallaga.