ANNOTATIONS AND ADDITIONS. 147 



I will only add the following remarks to this instructive 

 discussion by Professor Ideler. According to Pliny and 

 Solinus, Atlas rises from a sandy plain (e medio arenarum) ;. 

 and elephants (which certainly were never known in Tene- 

 rift'e) feed on its declivity. What we now term Atlas is a 

 long ridge. How came the Romans to recognise in this 

 long ridge the isolated conical mountain of Herodotus? 

 May not the reason be found in the optical delusion by 

 which every mountain chain seen in profile, in the prolon- 

 gation of its direction, has the appearance of a narrow cone ? 

 I have often seen in this manner, from the sea, the ends of 

 long chains or ridges, which might be taken for isolated 

 mountains. According to Host the Atlas is covered near 

 Morocco with perpetual snow, which implies an- elevation of 

 above 1800 toises, or 11510 English feet. It is also 

 remarkable that, according to Pliny, the " Barbarians/' i. e. 

 the ancient Mauritanians, called the Atlas " Dyris." The 

 chain of the Atlas is still called by the Arabs Daran, a word 

 which has almost the same consonants as Dyris. Hornius, 

 on the other hand (de Originibus Americanorum, p. 195), 

 thinks that he recognises the word Dyris in the Guanche 

 name of the Peak of Teneriffe, Aya-Dyrma. On the con- 

 nection between purely mythical ideas and geographical 

 traditions, and on the way in which the Titan Atlas gave 

 occasion to the image of a mountain supporting the heavens, 

 beyond the Pillars of Hercules, see Letronne's "Essai sur 

 les Idees cosmographiques qui se rattachent au nom d' Atlas," 

 in Eerussac's Bulletin universel des Sciences, Mars 1831, 

 p. 10. 



