ARMADILLOS — ANT-EATERS. 227 



studied on the spot the mammals both of Brazil and La Plata ; and in his works on these districts 

 he has enumerated the different species found in them, and out of twelve species of Armadillo 

 found in these two regions, he did not find one both in Brazil and La Plata.* The Dasypus 

 Peba, or nine-banded Armadillo, from Central America, has been stated to be an exception. 

 It occurs in Brazil, and in Guiana, and is said to be common in Mexico, and not uncommon 

 near the lower shores of the Rio Grande in Texas ; but it turns out that the Brazilian and the Texan 

 species are different, f I would preserve the name, D. novemcinctus, for the Brazilian species, 

 and Peba for the Texan. On the other hand, the great Ant-eater (Myrmecophaga jubata) is 

 found throughout the whole of South America, east of the Andes and down to La Plata, wherever 

 there is wood. No doubt it may be said that the physical condition of a wooded country is greatly 

 more homogeneous than one not entirely wooded, and less change is felt in passing from one part 

 of it to another. But the fact that there are onlj^ two species of Ant-eater as opposed to a dozen 

 of Armadillos, over the same extent of coimtry, shows, I think, that the one must be more 

 susceptible to the changes which induce modification of form than the other. 



The curious little mole-like Chlamydophouus truncatus, the Pichyciego, or blind Armarlillo of 

 the natives, is found in the neighbourhood of Mendoza ; but it does not extend across the Andes 

 into Chili, although Chili is usually given as its habitat. Neither it nor any other Armadillo occurs 

 on the western side of the Andes, until we reach Guj^aquil. Burmeister has lately found a second 

 species of this curious animal, also in that district, which he has named C. retusus.J 



The Armadillos have been divided into various sub-genera. Two sections seem sufficieutlj' 

 distinguished ; some having the fore-feet provided only with four toes, while others have the 

 fore-feet five-toed. But these distinctions do not seem to have any geographical import. The 

 Texan species (Dasypus Peba) belongs to the four-toed section ; but that section goes as far south as 

 the five-toed. Some of each being found in all parts of their range. 



III. Ant-eaters. (1.) — Manis. (Map 60.) The Pangolins inhabit the Indian Archipelago and 

 the land surrounding the Bay of Bengal, also the east coast of Africa from Sennaar to the Cape, and 

 the west coast from Senegal to the Niger. I do not find any record of its occurrence on the 

 south-west coast between the Cape and the Niger, its place there being seemingly occupied bj' the 

 Orycteropus, for which the arid, sandy coiratry is more adapted. 



FociUon, in a review of the genus (Manis)§ divides them into long-tailed and short-tailed 

 species. These have not much geographical significance, however ; the long-tailed, to be sure, is onlj- 

 found in Africa, but the short-tailed is found both in Africa and the East. 



(2.) Orycteropi. (Map 60.) Wholly African, and foimd on the west coast of Africa and South 

 of Nubia, near the White Nile. Owen observes that of all Edentata the Orycteropus most 

 nearly resembles the extinct genus Scelidotherium (one of the Megatheroid South American forms) 

 in the form of its cranium ; and next to it in this comparison, the great Armadillo (D.vsytus gigas.)|| 

 Although burrowers, and on that account classed by some with the Dasyi-i, tliey arc, as already 

 mentioned, entomophagous, and most nearly allied to the Ant-caters. 



* Burmeister, H., "Systematic Uebcrsiclit der thiere X Burmeister in " .Vbh. d. JJat. Gesellsch. zu Halle," 



Brazils," 1864. Burmeister, H., " Erlautening zur Fauna vii. 



Brasiliens," 1856. Burmeister, H., " Reise (lurch die La § Fociixon, Ad. in •• Rev. et Mag. de Zool." Sept. l>sr)0, 



Plata Staaten," 1861. p. 465. 



t Baird, on "U.S., Bac., R. R., Repub.," Vol. viii., || Owen, in " Zoology of the Beagle." 



624, 1S57. 



