54 SIDNEY H. RAY. 
particle appear as a separate word. But when, as in Tanna and 
Eromanga, the formative particle is represented by a single letter 
and cannot be written otherwise than in combination with the 
root, the inflectional character of the language is apparent. 
We may even compare the structure of a verbal form in Tanna 
with that found in the classical languages. Thus the singular 
imperfect indicative of the verb ‘to say’ is in Tanna: 
1. yak-am-ani 2. nuk-am-ani 3. t-am-ani 
and in Greek and Latin : 
l. &dn-y 2. &n-s 3. ey- 
1. dic-eba-m 2. dic-eba-s 3. dic-eba-t 
Here the Tanna personal prefixes yak, nuk, t correspond to the 
Greek suffixes y, s,¢ and the Latin m,s,¢. The Tanna tense sign 
am corresponds in meaning to the Greek augment ¢ and the Latin 
eba. The only difference in the three languages is in the position 
of the roots ani, dy anddic. The Tanna yak-am-ani, nuk-am-ani, 
t-am-ani are equivalent to ‘I, thou or he said,’ without the per- 
sonal pronouns tau, tk, te (demonstrative) being expressed, just as 
in the classical tongues. In all three languages however the 
pronouns may be used to emphasize the person. 
As also the Gaelic ba, ‘he was’ shows the Latin eba used as a 
separate word, so we find the Tanna am, represented by m (with 
varying vowel) used in more than one language of the New 
Hebrides as the sign of an action combined with a pronominal 
idea. In Malekula mz ry, and Espiritu Santo (Malo) mo viti are 
equivalent to ‘he speaks.” 
The descriptions of the highly inflected languages of the Aryan 
family given by Schlegel and Bopp are equally true of the Oceanic. 
Schlegel’s words are :—“ Le merveilleux artifice de ces langues 
(a inflexions) est, de former une immense variété de mots, et de 
marquer la liaison des idées que ces mots désignent, moyennant, 
un assez petit nombre de syllabes qui, considérées séparément, 
n’ont point de signification, mais qui déterminent avec précision 
le sens du mot auquel elles sont jointes.”* 
* Schlegel, quoted in Bopp—Comparative Grammar, p. 101. — 
