LANGUAGES OF THE INDO-EUROPEAN FAMILY. 405 



A mere glance at the natiu-e of the differences between the words 

 given above will suffice to show that physical conformation has 

 nothing, or at least little, to do with them, inasmuch as peoples who 

 reject the h, p,f or v in one case, keep it in the other. A survey of 

 the whole vocabulary of numerals tends to confirm this view. The 

 forms of the numeral ten may illustrate. In these, as in the forms of 

 eight, and as in the Coptic language to a very great extent, we find 

 the letters I and r interchanged. 



Earatonga nauru. Tonga ooloo orongoiooloo. 



Otaheite a-liooro. Tuham manud. 



Easter Island ana-hooroo. Sava bo. 



New Zealand aw^a-horro. Sandwich umi. 



Buges sopuloh. Philippine apalo. 



Paomotua hori-hovi Java sapoulo. 



Mo.rqicesas , ono-huu. Ne^o Guinea .saji^'a-foula. 



Madagascar fooloo. Samoa ti^ii. 



Batia sa-pooloo. Fiji wafulu. 



Mangavai puluh. Potti hulu.^ 



In this place I may also be permitted to allude to other forms of 

 the article, which have been so bound up with the substantive before 

 which they stand, or with the root to which their prefix gives a 

 substantive power, that they have been mistaken for part of the 

 root itself ; and thus the etymology of the words of which they form 

 part has been lost. The feminine form of the Coptic ai'ticle in T or 

 Th, which is supposed to have converted Ape, the head, into Tape 

 or Thebe, has, doubtless, some connection vdth the Hebrew feminine 

 termination, consisting of the same letter, or jl- Disregarding, 

 however, its feminine character, it would be the same as the Hebrew 

 n {i or th) abbreviated from j^ji^, the mark of the accusative and a 

 kind of article, which, prefixed to a verbal root, converts it into a 

 noun, e.g. LAMAD, learn ; TALMID, a learner. The language of 

 Lybia, or of the Shelluhs, differs fi'om that of the Canary Islanders 

 in many words by the possession of this prefix. Thus, temples in 

 Canarese are almogaren, and in Shelluh, tahnogaren ; a coarse article 

 of dress, called the haik, is, in the former, ahico, and in the latter, 

 tahajyh.^^ I do not imagine that every T or Th which can be shown 

 to be a prefix to the root, is a relic of an old article. In Hebrew, 



4* Mariner's Tonga Islands, by Dr. Martin. Edinburgh, 1827. Vocabulary. Labillardiere's 

 Account of a Voyage in Search of La Perouse. Translated. London, 1800. Vocabulary. 

 Bowring's Decimal System, 160—163. 



<5 Shabeeny's Timbuctoo, by Jackson ; Languages of Africa, 355—381. 



