पृष्ठम्:बृहद्देवता.djvu/३०

xxxiv] APPENDICES [Introduction parallel columns, the evidence for the relationship of the Brhaddevata to eight other texts, notably the Nirukta and the Sarvänukramani. The last and longest appendix (pp. 155-198) is a very full index of the words and names occurring in the Brhaddevata. I have excluded from it all words occurring in pratikas (Appendix i), in the list of authorities (Appendix ii), and the names of constantly recurring deities and the corre- sponding derivative adjectives (such as agni, agneya), as these may all be found by consulting Appendix iii. Thus, agni and agneya are omitted when occurring after the end of the introduction (ii. 125), unless there is some special reason for specifying them. Words of any importance found in various readings are also given and indicated as such (e.g. antra). I have found this index extremely useful to myself, not only for the purposes of the translation, but for the final revision of the text. The help afforded by the possibility of constantly comparing cognate passages has been very considerable, and has often led to certainty in regard to the correct reading or sense, when I should otherwise have remained in doubt, or perhaps altogether in the dark. Indeed, I doubt whether a difficult Sanskrit text can ever be satisfactorily edited without an index of the entire work made before the editor sends his first sheet to press. This is one of the reasons why I have long had a strong objection to the wretched method of bringing out texts in fasciculi. It has, as far as I can see, absolutely nothing in its favour. The editor probably begins to bring out his text before he knows it thoroughly, and the final fasciculus, if it ever sees the light, probably has, or ought to have, a long list of corrigenda in consequence. Meanwhile the work remains practically useless to scholars, while its interpretation is rendered additionally difficult by its hidden errors. Much loss of time is also entailed; for while, for instance, the whole of the present work of about 560 pages has taken less than a year to pass through the press, experience shows that four fasciculi of, say, 100 pages spread themselves over some six years in the process of publication. I am also convinced that it is the duty of every competent and careful editor, not only to refrain from this futile form of publication, but to extract as much material as possible from his text while engaged in editing it. As he must, necessarily, be more familiar with his text than any one else is likely to make himself, research is sure to be more rapidly advanced by the adoption of this method. 17. Acknowledgment of obligations. There now remains to me only the pleasant duty of thanking all those who have in any way assisted in the production of the present work. One or more valuable notes, which I have utilized in one form or another,