472 PROCEEDINGS OP SECTION G. 



quiry into the numeral systems now in use, will show that these are 

 founded on separate words for one, two, three, and no more, 

 and that the words for four and the higher numbers are in fact, 

 however much changed now, only combinations of these with one 

 another, and with words for h a n d, a n d, &c. 



IX. The number " five " presents some difiiculty. In many 

 languages it is the word for " h a n d," as Lepsius, Pictet, and 

 others have indicated. In most of the Malayan, Papuan, and 

 Polynesian dialects of Oceania, lima, " hand," is likewise the 

 word for "five." Now, although I do not assert that there is any 

 direct connection between the Keltic languages and the Oceanic 

 except a common ancestral origin, yet lima, " five," is very like 

 the I.-K. word 1 a m h (1 a m), the " hand," = that which "takes," 

 "the taker, holder," and this word I connect with the Gr. lam- 

 ban o, "I take," through the root g a m. The Scotch word h a d d, 

 "to hold," is of the same origin as the E. "hand." Some writers 

 connect the S. p a n c h a n, "five," with the S. p a n i, "hand," and 

 say that the pan is for p a m, as in the Aeolic Gr. p e m p e, 

 " five," and the Goth, f i m f, " five," C. pump, and that the pa 

 itself is k a , k a m, as in the L. q u i n-q u e. I have no doubt that 

 this root k a , or k a m, is the same as the root g a m found in 

 many languages in the sense of to "take," and in'L-K. in the form 

 of gabh ( = gamh), from which I derive G. 1 a m h, the "hand," 

 as "the takei\" The change of g into 1 is certainly uncommon ; 

 but the Gr. m o g i s and m o 1 i s are the same word, and the Aus- 

 tialian and Polynesian languages often cliange g into 1 ; the proper 

 name Lav-erna, " the goddess of robbers," finds no derivation 

 in Latin, but it becomes significant when we trace the Lav to 

 the Keltic root gabh, pi'onounced gav, " to take, to seize;" 

 and, further, the Gr. word 1 e - i s, " booty, " for 1 e (v) i (d) s, is to 

 me the same word as the G. ga-bh-aid, now contracted into 

 g a i d, "to steal." The change of g into 1 comes through the 

 intervention of d; thusg = d — 1. That the idea "hand" and 

 the idea "seize" are closely allied is seen in the S word 

 harana, the " hand," which, as an adjective, means "taking, 

 seizing," and its S. root har-y is for hri, originally bhri, 

 " to take." Again, the Gr. word c heir, the "hand," altliough 

 Curtius gives a different derivation for it, seems to me to come 

 from the same root gabh; for ga-bh-air, with the b li 

 silent as is usual in .such a position, closely approximates to 

 c h e i r . Be that as it may, the G r. p u x , p u g -me, the 

 "fist," is the same word as the G.-Ir. kaog, koig, kuig, 

 (p for k), "five," and this word kaog, as I think, ought to be 

 assimilated to lamh, the "hand," by writing it ka-mh-ag or 

 ka-bh-ag (bh, mh silent), so as to show its relation to the 

 verb gab h, which gives also the L. c a p - i o , "to take, to seize." 

 Kuig becomes L. q u i n q u e by giving to g the nasal sound, 



