684 ANNUAL EEPORT SMITHSONIAN INSTITUTION, 1912. 



he struck it with his right hand, and variously modified the sound 

 by all sorts of muffling manoeuvers with his left. Witte has recently 

 thorouglily investigated the "drum speech" of the Ewe tribe and 

 reported on it in the "Anthropos" (Wien, 1910, Heft I). From his 

 very clear statements, it presupposes a power of understanding 

 whose development does credit to the ear of primitive peoples, but 

 does not involve an appreciation of art. The percussion instrument 

 most capable of producing melody is the famous marimba of the 

 Kaffirs, a kind of xylophone made over hollow gourds as soundmg 

 boards. The metal wind instruments remained entirely undeveloped. 

 They were only of value for signals. Moreover, most of the so-called 

 trumpets served jirincipally, not to produce tone, but only to increase 

 sound; they were to be shouted into like a speaking tube. The 

 stringed instruments stand on a somewhat higher level though we 

 find forms similar to the harp, guitar, and zither. 



Certainly all the stringed instruments are later than the percus- 

 sion instruments. The fact has already been demonstrated that none 

 of the finer melodic instruments have any religious significance. The 

 only inusical instruments which by tradition or use are exceptions 

 are the men's flutes of the Papuans of New Guinea (whose tone was 

 regulated by a sliding rod); the rain pipe of the Basutos, a kind of 

 horn made out of bark by the Indians of Rio dos Uaupes, one of the 

 tributaries of the Eio Negro (women who obtain sight of it are ruth- 

 lessly poisoned); and a shell trumpet used by the priests of the Fiji 

 Islands in sacred processions. None of these instruments, however, 

 in the manner in which they are used produce melody. All melodically 

 used instruments, on the contrary, have neither a magical nor particu- 

 larly a religious character. They are secular, which fact certainly 

 indicates their late origin. 



It is necessary now to establish the origin of the different instru- 

 ments. The most obvious and formerly most commonly accepted 

 view was that prmiitive peoples had independently contrived the 

 whole of their orchestra, and had perfected it from rude beginnings. 

 This supposition is, however, strongly shaken by more recent inves- 

 tigations. Thus, it has been shown in the case of the most interesting 

 melodic instruments that they have not resulted through development, 

 but are rather retrogressions from more perfect borrowed examples. 

 The negro harp boasts its origin from the corresponding old Eg}^tian 

 instrument; the pipes of the Kubus in Sumatra are the product of 

 Javanese civilization (Hornbostel) ; the marimba imitates Chinese and 

 Indian predecessors (Wallaschek) ; and the lute came first to Africa 

 with the Arabian invasion (Ankermann). 



It is now in the highest degree imperative to observe in what 

 direction the retrogression moves. The workmanship becomes care- 

 less, the contours of the instruments are less shai^ply defined, a poorer 



