52 



NA TURE 



{May 20, 1 8 So 



language. But it seems obvious that this diversity and 

 complexity must ha\-e been evolved in the natural course, 

 whether starting from one or many original centres. 



At p. 163 a view is taken of the Ai7an suffixes which 

 many will be inclined to regard as a retrogressive step 

 rather than an advance in linguistic studies. "We must 

 rid ourselves of the notion that suffixes were ever inde- 

 pendent words like our 'if or 'in'; so far back as our 

 knowledge of Aryan speech extends they possessed no 

 existence apart from the words to which they belonged, 

 and which, again, only existed as words in so far as they 

 possessed these suffixes. Suffixes became flexions through 

 the help of analogy." The point would involve too much 

 technical matter to be here adequately discussed, but it 

 may be remarked that our knowledge of Aryan speech is 

 as of yesterday compared with the many]^ages it must 

 have taken to reach the highly-inflected state presented 

 by the oldest known members of the family. If in a 

 brief thousand years or thereabouts the Latin ablative 

 meiit: had time to become a Romance adverbial suffix, 

 the verb habco a verbal ending, and the adverb indc a 

 pronoun with a genitival force, surely there was ample 

 time in the ten, twenty, or fifty thousand years of the early 

 lifetime of the organic Aryan speech for hundreds of 

 independent words to pass from one part of speech to 

 another, from the noun or verb to the particle, and thence 

 to the relational suffix. And if "suffixes became flections 

 through the help of analogy," being hitherto " meaning- 

 less terminations " {ib), it maybe asked through the help 

 of what analogy ? At all events, the internal vowel change 

 here taken as their pattern does not meet the case, for, if 

 properly considered, all such internal vowel change must 

 itself be regarded as primarily due to the influence of 

 reduplication and flection acting on the body of the word, 

 and gradually becoming absorbed, often leaving no trace 

 of its former presence beyond the very vowel change in 

 question. Such seems undoubtedly to be the history of 

 the strong Teutonic conjugation and of such Teutonic 

 plurals as seem now to be effected by mere internal modi- 

 fication, just as we know that it is the history of such past 

 tenses in Latin as cgi, feci. Two things it seems im- 

 possible to admit — the development or invention of 

 "meaningless terminations," that is, meaningless ab 

 initio, and internal vowel change with flectional force, 

 produced, as it were, by spontaneous effort independently 

 of outward influence, the influence either of reduplication 

 or of pre- or postfixes reacting on the theme. 



The chapter on Comparative Mythology, as expounded 

 in the light of comparative philology, is thoroughly satis- 

 factory, and will be read with pleasure even by those un- 

 familiar with the technicalities of the subject. In the 

 last chapter, also, on the Origin of Language and col- 

 lateral subjects, much excellent advice is given touching 

 spelling reform, the pronunciation of the classical tongues, 

 the application of sound linguistic principles to the 

 teaching of languages, and many other points of a more 

 practical nature. 



There is an excellent analytical index supplied by Mr. 

 W. G. Hird, but it does not dispense with the necessity 

 of a full alphabetical index, which is urgently needed in a 

 work overflowing with matter of the most varied descrip- 

 tion, and which it may be hoped will be supplied in future 

 editions. Some oversights and casual slips in minor 



points should then also be rectified, and with that view a 

 few of the more important may here be noted. The ve 

 in the Italian compound portandovclo (ii. 210) is derived 

 from the Latin adverb ibi, used pronominally instead of 

 from the pronoun vobis. The particle vi, ve often, of 

 course, represents ibi, as in the sentence io v'era (lit. ego 

 ibi eram) ; but it equally represents the pronoun, as in 

 the sentence io vi dico (lit. ego vobis died), and obviously 

 in the compound in question. The Nogairs (properly 

 \ogais) are described (ii. 199) as "Russian Cossacks" 

 instead of Tatars. The Nogais are of Tiirki stock, 

 whereas all the Cossacks are of Slav stock, either Great 

 Russians (Don Valley, Cis-Caucasia, &c.), or Little 

 Russians (Ukrania). The Cossacks are often spoken of 

 as Tatars by careless writers, confounding them with the 

 Cassaks, who, being Kirghizes, are true Tatars. It seems 

 scarcely accurate,^ to ^say that in the Greek and Latin 

 sentences rvTrrei and amat " the subject is not expressed" 

 (ii. 329), seeing that d (for eVi) and the / of amat arc pro- 

 nominal, though so old that they do not distinguish the 

 gender of the^ subject referred to, and may possibly have 

 originally been objective forms. The statement (i. 417) 

 that " in Hindustani the genitive takes the marks of 

 gender according to the words to which it refers," is apt 

 to mislead the unwary, who might conclude from this that 

 the Hindustani noun had cases, whereas there is nothing 

 but a general oblique form followed by postpositions. 

 One of these postpositions {ka = of) follows the gender 

 of the noun of reference {larke-ka, larke-ki according to 

 circumstances), but the noun remains unchanged. There 

 is another reference (p. 423) to a point of Hindustani 

 grammar, which as worded is unintelligible. The place 

 of the definite article is not supplied " by a dative with 

 the suffix -ko^' for there are no datives, but by the post- 

 position ko, which, though usually giving a dative force, 

 often idiomatically emphasises the objective noun and 

 thus does duty as a sort of definite article. The reference 

 to Voltaire (i. 60) should be emended by shifting the 

 places of the words "consonants" and "vowels." No 

 one who has ever heard a native of Northern India speak 

 any of the current neo-Sanskritic tongues will hesitate to 

 transcribe the sonant explosives with the rough breathing 

 {gha, dha, bhd) by the side of kha, tha, pha, though the 

 point is treated as doubtful (i. 281). The h in such words 

 Asghora, bha,l, dkobi is heard quite as distinctly as it is 

 in the English word mad-house. Lastly, such terms as 

 "Turanian" (i. 325), "Alfurian," and even Malayo- 

 Polynesian might well be dispensed with in future editions 

 of a work, which as it stands reflects lasting credit on 

 English scholarship, and Avhich all will accordingly be 

 anxious to see rendered even in small details as perfect as 

 possible. A. H. Keane 



STATICS 



Treatise on Statics. By George Minchin, M.A. Second 



Edition. (Clarendon Press Series.) 



SINCE the publication of Thomson and Tait's " Natural 

 Philosophy," thirteen years ago, an important change 

 in the treatment of the theory of dynamics has been 

 making rapid progress. Previous to that time it was the 

 almost universal practice to follow the French writers and 

 to find a basis for the theory of the equilibrium of forces 



