432 Report S.A.A. Advancement of Science. 



common in most languages. The constructive uses which such 

 changes may be put to in language we shall note presently. 



2. We may note also the frequent occurrence of Iteration which 

 aflfects only parts of words. Under this head we might again sub- 

 divide into Iteration at the beginning, at the middle, at the end of 

 words, as well as the combination of these partial iterations with 

 vowel changes. A further subdivision might be made according as 

 the iteration is of the vowel or consonant part of a word, but we 

 need not elaborate overmuch. The phenomena of such partial 

 iterations are most commonly found at the beginnings or ends of 

 words, and are perhaps more conspicuously seen when they affect 

 consonants. 



Ex. : — a. Momordi, pepuVi. 



h. Alig//g//li (from the singular alguli). 



C. ipVKHKU) {ipVKii)), ike, &C. 



III. Thirdly, we may note what may be called veiled Iteration 

 — i.e., cases where there is fair reason to conclude that certain 

 phenomena of language are due to- an original Iteration which is no 

 longer actually visible, but which has left traces of its effects on the 

 structure of words. Here, of course, we are on much more unsafe 

 ground, and in many cases cannot be sure whether the phenomena 

 are due tO' the residue effects of iteration or to some other cause of 

 phonetic change. For example, when we find in Latin an iterated 

 or reduplicated form like pepcri, and find that a compound of this 

 verb repperi has dropped the visible doubling but has retained a 

 double p, it is obviously fair to conclude that we have in this con- 

 sonant doubling the veiled form of iteration. But when we find that 

 certain verbs in Latin form perfect stems by a vowel lengthening, and 

 that in Greek a similar lengthening is accompanied by Iteration, it 

 is not so certain a conclusion that we have in this lengthening a 

 veiled reduplication, for we find other indications that such a 

 lengthening may be due to other phonetic causes. Still, however 

 uncertain we may be as regards particular cases, there can be no 

 doubt that we have here a vera causa and that if we had the means 

 of tracing the history of dropped iterations we should see their 

 effects in many of the phenomena of phonetic change. 



We may now look at the various ways in which iterative forms 

 are used in language, and we shall see that Iteration plays an im- 

 portant part, whether language be in its most primitive or in its most 

 highly developed forms. We shall look at the various uses of 

 Iteration in what may be regarded as a kind of chronological order, 

 though it must be remembered that the earliest uses survive to> a 

 certain extent and are to be seen side by side with the later. We 

 should naturally expect that this would be the case, for man, how- 

 ever highly developed, is still an animal, and Iteration may be said to 

 be based on an impulse common to all animal nature. " Naturam 

 expellas furca tamen usque recurret " — and " Scratch the Russian 

 and you will find the Tartar " — are proverbial expressions which find 

 their exemplification in language continually. The (H)arr}- who 



