TRANSACTIONS OP THE SECTIONS. 157 



original family from which the Indo-European races have sprung. This fact may be 

 inferred from the word being used by remote subfamilies ; but its being used by two 

 subfamilies (or even by a single one), and being also found in one of the languages 

 that are cognate to the original Indo-European, but not derived from it, is still 

 stronger evidence. Thus the word horse, which is peculiar to the Teutonic family, but 

 which is cognate to the Latin enrro, " I go like a horse, i. e. I run," is shown to 

 have been in use in the original Indo-European family from its evident connexion 

 with the Accadian kurra. The original Indo-European root must have been kurs. 



2. Such a comparison may determine the original form of an Indo-European root, 

 which varies in the different languages known to us ; as it may also determine the 

 original Semitic form. For example, in p. Ul, Report for 1857, the facts that the 

 original Indo-European form of the second numeral was ihwi and the original Semitic 

 form thni, could only be established by a comparison of the different forms known to 

 exist with the Accadian mi, as explained in that paper. In treating of the words for 

 "lion," No. 14, he was ignorant that the Assyrian word was libbii. Comparing 

 this with the Accadian lig, he now thinks that the Semitic form must have been ligb, 

 and that this was also the Japhetic form. The Indo-European root would be ligw. 

 In Latin this would be declined lix, livis ; and, as (s)nix, (s)nivis gives snow in 

 English, and snig, sneg in the Letto-Sclavonian languages, so liiwe in modern and 

 lew in older German correspond to lix; and lig might be the ancient Letto-Sclavo- 

 nian root, equivalent to these. 



3. An etymological relation between Indo-European words may possibly be 

 established if both can be shown to correspond to the same word in a language 

 which is not Indo-European. Thus, the relation between yXuxuj and ykao-cra is not 

 generally admitted by Greek lexicographers ; though if the words be written as they 

 would be in the Cadmean alphabet (see p. 142 of the former paper), yXotc-Fes and 

 ykoK-ia, the resemblance is easily seen. But this relationship is established when 

 we find that ghlu is used in Egyptian both for " sweet " and for " tongue." The 

 Latin dulcis, originally dlucvis, is cognate to these. 



Enough, however, on the subject of the Accadian words occurring in the Assyrian 

 inscriptions. They present no ethnological difficulty, as the people who spoke the 

 language to which they belonged lived close to Assyria. The case is the same with 

 the Semitic words which occur in the Egyptian inscriptions. He pointed out one in 

 1845 ; rDDID " a chariot ; " used also for " chariots ; " the Egyptians not generally 

 expressing the vowels, and thus writing the terminations eth and 6th alike. Mr. 

 Birch has since found another meaning, " round bucklers," which appears to be the 



Arabic .x& JLc. These are Canaanitish words, expressing objects brought from 



Canaan. 



There were words, however, in the Assyrian inscriptions which Dr. Hincks be- 

 lieved to be Indo-European; and as no Indo-European people had been hitherto 

 recognized as existing on the west of Assyria, their existence presents an ethnolo- 

 gical difficulty which the author seeks to explain in this paper. 



The words in question were ligwindinas and lasanan. The former occurs on a 

 great slab or altar in the north-west palace at Nimrud (B. M. Series, pi. 44, 1. 17). 

 Every other word in the sentence is of known signification. " L. alive I took cap- 

 tive." From the context it must necessarily be the name of an animal, in the plural 

 number (being joined to a plural adjective) and in the accusative case (being governed 

 by a transitive verb). Every one who has the slightest knowledge of the grammar 

 of the Semitic languages must see that ligwindinas cannot be a Semitic accusative 

 plural ; and every one that knows anything of the grammar of the Indo-European 

 languages must see that it is a regularly formed Indo-European accusative plural. 

 Taking it as Sanskrit or Zend, the nominative singular would be ligwindi; taking 

 it as Greek, it would be ligwindis or ligwindin. Whatever be the particular language 

 to which it belongs, and whatever be the meaning, its being Indo-European ought 

 not to be questioned. The other word, lasanan, has puzzled the interpreters of the 

 Assyrian inscriptions as much as any other word. It occurs in several contexts : 

 '* king lasanan " after " king of Assyria " on Bellino's cylinder, 1.1; " ruler of the 

 tribes lasanan," on the Tiglath Pileser cylinder (I. 29), " wielder of the sceptre lasa~ 

 nan," same cylinder (VI. 56). "Assur the great lord has mademetoyossMs (? yusad- 

 limanni ; the first radical may be T, O or D) the kingdom lasanan," Bellino's cylinder, 



