plural-suffix also among the plural endings of our earliest nouns, but with only faint and doubtful success; if these are actually of composite derivation, the marks of their composition are hidden almost beyond hope of discovery. We must be content to say for the present, at least, that the suffixes of declension indicate by their differences the distinctions of number as well as of case. And, among the nouns as well as the verbs of the primitive language, not only a plural, but also a dual, was distinguished from the singular by its appropriate endings, which are of not less problematical derivation, and, in the earliest condition of speech that we can trace, much fewer in number, being limited to three.
One other distinction, that of gender, was partially dependent for its designation upon the case-endings. We have already (in the third lecture) had occasion to refer to the universal classification of objects named, by the earliest language-makers of our family, according to gender, as masculine, feminine, or neuter—a classification only partially depending upon the actual possession of sexual qualities, and exhibiting, in the modern dialects which have retained it, an aspect of almost utter and hopeless arbitrariness. Nor, as was before remarked, is it possible even in the oldest Indo-European tongues to trace and point out otherwise than most dimly and imperfectly the analogies, apparent or fanciful, which have determined the grammatical gender of the different words and classes of words: such is the difficulty and obscurity of the subject that we must avoid here entering into any details respecting it. It appears that, in the first place, from the masculine, as the fundamental form, certain words were distinguished as possessed of feminine qualities, and marked by a difference of derivative ending, often consisting in a prolongation of the final vowel of the ending; while to all the derivatives formed by certain endings like qualities were attributed. The distinction was doubtless made in the beginning by the endings of derivation alone, those of case having no share in it; but it passed over to some extent into those of case also, the feminine here again showing a tendency to broader and fuller forms.
18
