only necessary to trace the German selecht back some centuries to find that it has changed from its original to a meaning diametrically opposite. Slecht (good) meant formerly right, straight, but also simple, which presently came to mean foolish or useless, and useless, bad. The Latin ingenium, which in Latin signified inborn faculty, became degraded in Italian to ingannare, to cheat. Selig, in Anglo-Saxon blessedbeatus-appears in English as silly.

Highly interesting, indeed, is it to watch the changes in form and meaning displayed by words passing from the Ganges or Tiber to the sea of modern European speech. Latin, it is unnecessary to say, which in the eighth century B.C. was the dialect of a small Italian territory, from becoming the tongue of the conquering Romans not only assumed a dominance in Italy, but being the language of law and government asserted a prominence in a great part of Europe; and when coming in contact with the vigorous idioms of the Teutons, though it could not supplant, it left a layer of its vocabulary on their language, so that common words once spoken by the Italian shepherds are now in use among the philosophers of Germany, the poets of France, and statesmen of England. Take the word palace; one of the seven hills of Rome, the Palatine, was dedicated to the pastoral deity Pales, whose festival, celebrated on the 21st April, commemorated the day on which the wolf suckled Romulus drew the first furrow for the foundation of Rome. In imperial times Nero built his golden house, which was hence called Palatium, on this hill, and the word thus came to be applied to the palaces of the kings and emperors of Europe. The English word court, French cour, Italian corte, has a similar strange etymology. The enclosures or cattle yards on the hills of Latium were called cohors, a word subsequently applied to the divisions of the Roman army. This word, which was used in Rome in both a pastoral and military sense, with a difference of pronunciation merely, became curtis; and was applied in medieval Latin to farms and castles of the Roman settlers, which became the centres of villages (hence the modern

names Vrau court, Graincourt, &c.); and lastly, from a fortified place rose to the dignity of a royal residence. Many other words have as long and surprising a pedigree. Thus the English lord, in the sense of nobleman, was in Anglo-Saxon half-ord, the warder of bread; and lady, in A Shlæfdige, she who looks after the loaf; earl from the Danish Jarl, elder, hence alderman; count, from comes, a companion; baron, the mediæval baro, man and knight; the German knecht, servant.

Professor Muller reiterates in his first lecture the principle which formed the basal thesis of a department of his previous volumethat thought, in the sense of reasoning is impossible without language. Infants and animals have each memory, sensation, perception, and instinctive judgment, but no trace of the faculty named by the Greeks logos-both ratio and oratio-a word derived from legein, to gather, collect, and classify ideas, to select the single from the general. The mind would be as incapable of recounting things without words, as of counting quantities without numerals. Man could not give nomenclature to a tree, animal, or river, without discerning in each some general characteristic quality; and though the savage in his earliest state may have named a horse by imitating his neighing, such sounds are not language, nor is it in this way that words are formed. Thus the Arian word for horse has no resemblance to neighing; derived as it is from the root as, it represents the quality peculiarly recognised in the animal. So wheat was not named from its being bearded and waving, &c., but simply white, to distinguish it from other edible productions. Roots are those words in every language which cannot be reduced to a simpler form; constituting the last residuum to which analysis can reduce the dialects of an inflexional tongue, they comprise the original germs of human speech. At the present day the world presents several languages which have remained in this germinal condition. The Chinese, for example, in which a monosyllable is at once noun, verb, and particle, its meaning depending on its place in the sentence; and the same is the case with the Polynesian

dialects and with the ancient Egyptian, as proved by Bunsen.* We need not here recapitulate Muller's theory that every language has passed through a radical and agglutinative to an inflexional stage, or that originally the constituent elements of words, of which, from the action of phonetic change and dialectical regeneration, often no more than a letter remains, were once the real symbels of thought and speech. The attempt at this period to discover the beginning of language must remain an insoluble problem, as illustrated from the light thrown upon it by the history of its later changes; though aided by imagination, physiology may yet indicate the nature of the first sounds spoken by savage man, as derived from the impressions of the five senses; the number of representative words formed increasing with the progress of the social state. Nothing can be more rational, even though devoid of the absolute proof afforded by the tongues of the oldest civilized and existing savage peoples, however, than the supposition that the earliest body of all languages consisted of monosyllables, and that the progress to more intellectual systems of expression resulted from intercommunication of races separated by locality, and the mental necessity and development of aggregates in the primary conditions of civic life.

After all that is formal or the result of grammatical art has been removed from language the radical element remains; and all that can be asserted with respect to those of the Arian languages is, that they have d'finite forms and mean r2s, nothing being now ascertainable of the monesy labre or a zgutiative stazes through which the dialect which resulted in Sanscrit and its relational branches passed. Such roots have universally a typical meaning, and as regards their origin are not resolvable by any interjectional or onomatopeie principle. Even the earliest Greek philosophers were aware that it was im

possible to trace word making to a mere imitation of sounds. Heraclitus, in his enigmatic manner, said that words were the shadows of things, meaning that man assigned such names to the objects around him as he considered expressive. Democritus represented language as due to thesis, institution, or envention; he does not regard words as natural images, but works of art, "statues of sound." Vague exposi tions enough, but sufficient to show that the earliest Greeks by no means considered language as an imitation of sounds. Speaking of the formation of language, Muller, adopting the principle of natural selection, proceeds to say that, though a series of sensuous impressions produce a mental image, and a number of them a general notion expressed in a vocal cry, the faculty of reason in man ultimately resolved such constantly recurring and most useful sounds into a root expressive of the general notion of their objects. Subsequently new combinations and eliminations occurred, in which process the new word was determined by its appropriateness and utility.

In proceeding to show that all names are general terms, Mulier produces a number of interesting ilustrations from various languages. Some nations are without names for the numerals beyond tour, ali beyond this being aggregated under the idea of many. In the Hawaian dialects there are no terms to express the difference between black, blue, darkness; between love, friendship, giat.tude, benevolence, &c. The northern Turanian races have no word for river, though many for the smallest rivulet; no word for tree, but many for birch, fir, ash; but even these special names are really general terms. In the Society Islands, when dogs were first introduced, the natives eed them brooas, their name for pigs; and the same thing occurred at Tima, instead of giving them a name from their barking, a singular example

of the origin of terms, and quite opposed to the onomatopic theory. Subsequent observation of the qualities of the animal possibly generated many names, which natural selection finally reduced to one. Thus in Sanscrit there are examples of clusters of roots expressive of a common idea. That selected by the Professor is the root Mar, of whose history in its passage through the world he gives an interesting account, but one too long to enter upon.

In treating of the importance of metaphor in the construction of language, Muller quotes Locke, to the effect that all metaphorical expressions were originally derived from the terms of objects of sense; but shows that his principle of the primary existence of a series of arbitrary articulate sounds to signify definite ideas is unsupported by evidence. Metaphor is the transference of a name from an object to other objects, in which the mind traces some participation of relations attaching thereto. Most roots had a material meaning originally, and one so general as to be applicable to special objects. Thus from those meaning to shine were formed sun, moon, eyes, gold, silver, happiness, love; for those meaning to go--clouds, ivy-creepers, serpents, cattle, chattle property; for those meaning to crumble-sickness, death, evening and night, old age, the fall of the year. There are two sorts of metaphor-radical and poetical; the first, when a root expressive of shining is made to form names for sun, spring, brightness of thought, &c.; the second, where a noun or verb is transferred to another object or action, as when the sun's rays are called its hands, clouds mountains, lightning an arrow, &c. That there was a mythical period in the history of the human race, when all thoughts that referred to matters beyond the narrow circle of every-day life were expressed in metaphor, is, indeed, not only true, but equally so that the mental relational action, of which the mythologies are an exponent, is one of the phenomena of current existence. Riksha, a Sanscrit word, derived from a root to be bright, and having an homonyme signifying bear, was the name given to the constellation of the ursa major. It meant

originally the seven bright ones; and afterwards, by a verbal confusion, came to mean the seven bears, possibly from some resemblance to the animal. Without understanding the origin of the name the Greeks and Romans called the constellation arktos and ursa, and thus the term arctic regions (arctus-ouros) has its rise, from a misunderstanding of a term made ages ago in Central Asia. When the Greeks spoke of the stars as the eyes of the night, of Argus, the all seeing (panoptes), and proceeded to describe his body as covered with stars, we have an example of poetic metaphor, and one of mythological from verbal application.

In a very amusing essay on Popular Etymology, Muller gives many illustrations of the singular and contradictory meanings which have been assigned to words and expressions. From the Anglo-Saxon blot, sacrifice, blotan, to kill for sacrifice, was derived blessian, to consecrate, to bless. The signs of many old taverns are instances of hieroglyphic mythology, such as "A Cat with a Wheel," a corruption of "St. Catherine's Wheel;" the "Bull and Gate," originally a trophy of the taking of Boulogne Gate by Henry VIII.; and the "Goat and Compasses," a corruption of the Puritan signboard, "God encompasseth us.' In Lincon there is an ancient gateway called the Grecian Stairs. These were originally called Greesen, the old English plural of gree stair. When the meaning of greesen was lost, stairs was added thereto, and the title ultimately changed as above. The proverb quoted by Hamlet, "to know a hawk from a handsaw," was originally "to know a bawk from a hernshaw," a sort of heron. The most singular instance of popular etymology adduced, however, is that of the barnacle goose. There is a species of shellfish named barnacle, which attaches itself to floating timber, and whose neck exhibits some remote resemblance to feathers. There is also a goose called bernicula, common to the north of Scotland and Ireland. The similarity of the name and fancied likeness just referred to, gave rise to the myth that the barnacle goose originated from the mollusc. Several curious passages are adduced from

Chinese, Mexican, Peruvian, or s in the Australian; even Sanscrit has nof or soft silibants; Greek no y, w, or ƒ; nor Latin soft silibants or aspirates like the Greek 9,9, x; English, no guttural breaths like the German ach, ch. R is a letter which some nations find it impossible to pronounce; thus, instead ⚫ of Christ the Chinese say Ki-li sse-tu; Eulopa for Europe, Ya-me-li-ka for America. Lis unknown in Zend, the Cuniform inscriptions, and various American and African tongues. It is curious to consider the absence and presence of certain letters in certain languages, the cause of which forms an interesting problem for the philosopher and physiologist, who have yet to determine how far such diversities are the result of structure, climate, or habit. As regards consonant sounds, Hindustani possesses 48; Sanscrit, 37; Turkish and Persian, 32 and 25; Arabic, 29; the Kathir, 26; Hebrew, 23; English, 20; Greek, and Latin, and Mongolian, 17'; Finnish, 11; Polynesian, 10; Australian, 8. From this list it is evident that the tendency to pronounce consonants is not dependant on varieties of temperature or social progress among the earth's races. Of the inability of different people to distinguish between certain consonants many singular examples are given in this lecture.

Professor Muller enters into the question of the changes in the same word which are found in dialects of the same tongue, and regards such phonetic alterations as attributable to the action of a regular law-a fact which he illustrates from what has been observed to occur in the Kaffir dialects; while the excision of letters seen in many English words, compared with the original Anglo-Saxon, such as lord for hlaford, lady for hlæfdgé, he attributes to the effect of laziness, an attempt to economize breath, and want of muscular energy. Other changes of an opposite character, from such phonetic alterations, which manifest an equality of vocal power, he names dialectic growth; and accounts for the literal changes exhibited in the Latin by those formed by the Roman soldiers settled in Dacia and still observable in the modern Wallachians, such as p for q, on the supposition that the original colonizers were Oscans and Umbrians, this pe

VOL. LXV. NO CCCLXXAV

culiarity being seen in their inscriptions.

The fifth lecture is devoted to an exposition of the famous law of Grimm-a discovery which enthusiastic philologists have placed on an equality with those of Kepler-a law which illustrates the fact, that similar roots are found in the IndoEuropean group of languages. Wherever the Hindus and Greeks pronounce an aspirate, the Gotns, Low Germans, Saxons, Anglo-Saxons, Frisians pronounce the corresponding soft check; in other words, that the Greek and Sanscrit kh, th, ph, becomes g, d, b in Gothic; and k, t, b in old high German. This phonetic law is only discoverable in highly developed languages, and is not referable to those in a primitive or secondary condition-monosyllabic or agglutinative.

One of the most valuable disquisitions in the volume is on the mythology of the Greeks, lecture 9. The acceptance of a mythology so monstrous as that of the Greeks, by a people possessed of such supreme intellectual gifts, has long remained one of the speculative puzzles of history. The perfection of the Deity formed the conceptive basis of his existence with the earliest thinkers, as with us; yet the stories of Kronos, Demeter, Tantalus, &c., would hardly be accepted into the pantheon of the lowest savage. Zenophanes, the precursor of Pythagoras, remarked that man had created his gods in his own image; and Heraclitus laughed at the Homeric theology, Anaxagoras interpreted the legends of the gods allegorically. Socrates was a sceptic, and many of the poets and philosophers of a later period exposed the inconsistency involved in such deitific ideals. To a logical being like the Greek such conceptions must have been utterly irrational; and those, indeed, who occupied themselves with their mythology gave it three interpretations- namely, ethical, or that which supposed the tales of the gods, their power of rewarding and punishing, was an invention of the wise of old; physical, such as that of Epicharnus, who declared that the gods were the elements; and historical, like that of Euphemerus, who resolved such myths back to the lives and actions of an

3