OFTHEIIUMANFAMILY 39 



father's side or on the mother's side. Naptar and naptrl are restricted to grand- 

 son and grand-daughter, although, without much doubt, they were originally applied 

 to a nephew and niece as well. From the diagram it is a proper inference that the 

 remaining persons in the several lines are described in a similar manner. The 

 Sanskrit system appears to agree with the general form prevalent in the Aryan 

 family. In its development it took the same direction before noticed in the Grecian, 

 and, to a great extent, in the other dialects of the Aryan language, but without 

 changing essentially its original form.' 



This term, together with that of aunt from amita, has been adopted with dialectical changes into 

 several of- the branches of the Aryan family, and promises ultimately to displace indigenous terms 

 dereloped since the separation of its branches from each other. 



In the-order of time a term for cousin would be the last invented, on the supposition of a growth 

 of the nomenclature outward from Ego. It is the most remote collateral relationship discriminated 

 in any language or dialect represented in the tables, unless the Slavonic is regarded as an exception. 

 A special term for this relationship must be founded upon a generalization of four different classes 

 of persons into one class; and, therefore, it is more difficult than either of those previously named. 

 This term cousin, which seems to be from the Latin consobrinus, was in strictness limited to the 

 children of sisters ; but it became a common term, and from this source it has been propagated into 

 several branches of the Aryan family. With these facts before the niiind it becomes more and more 

 apparent that the original system of the family as to its present form was purely descriptive. 



' Note on Sanskrit- Schedule by Fitz Edward Hall, D. C. L. : — 



1. The prescribed scheme of vowel-sounds being very inadequate for the Sanskrit, I have 

 adhered to that more usually followed by Orientalists. According thereto, a is like a in "father;" 

 a, like a in " America ;" e, like our alphabetic a ; i, like 1 in " pin ;" i, like i in " machine ;" 0, like 

 in "no ;" u, like u in " bull ;" u, like 00 in " fool ;" ai and au, as in the Italian. A peculiar vowel 

 is represented by ri, which is sounded somewhat like the ri in "rivalry." Sli, s', and s, indicate 

 three different sibilants. 



2. In consequence of prefixing mama, " my," to each word, I have had to give it a case. I have 

 selected the nominative. The crude form, that found in the dictionaries, of the words for "father," 

 "mother," "son," "brother," &c., are pitri, rndtri, bhratri, putra, &c. 



3. It requires great credulity to believe that the Hindus know much of the origin of Sanskrit 

 words. Generallj', they can only refer words to verbal themes, which are, of course, the invention 

 of the grammarians. Putra, " son," for instance, is fancifully derived from pu, one of the " hells," 

 and the etymon "tra," "to draw out;" quasi, "an extractor from hell." Duhitri, " daughter," is 

 thought, with more of reason, to mean "the milker." See Prof. Mas Miiller on Comparative 

 Mythology, in the Oxford Essays. Pautra, "grandson," is from putra, "son." To paulra, the 

 preposition pra, "before," is prefixed in prapautra, "great-grandson." "Elder brother" and 

 "younger brother," agraja and anuja, mean, when analyzed, " foreborn" and "after-born." In 

 pitdmaha and mdtamaha, "paternal grandfather" and "maternal grandfather," and so of the femi- 

 nines, maha and mahi are inseparable affixes. The vriddha, in the word for "great-great-grandfather," 

 imports "old." Pati, " husband," " lord," we have in the post-Homeric Sfondrijs, the first syllable 

 of which is the same as the Sanskrit drsa, "country." The feminine of pati, patui, occurs in the 

 Homeric and later iianoiva. Dhava, " husband," is seen in the Latin vidua, in Sanskrit, vidhavd, 

 "without husband." Hence appears the absurdity of the masculine viduiis, and so of our "widower." 

 Vimdtri, "step-mother," moans "a different mother;" for vi has numerous senses in Sanskrit. 

 Dattaka, " adopted son," = " given." In vimdtreya, " half-brother," we see vi and vidt7-i, " mother." 



4. Degrees of relationship representable only by compounds of other degrees have been omitted. 

 And here I should mention that intrivtja, "father's brother," is the only word for "paternal uncle" 

 in Sanskrit. It contains jnlri, " father," and an ending. Compare tjlirulrivija and hhdgineya. 

 Ildtula is connected, not very obviously, with mdlri. 



