chiefly instrumental in making it popular, say in their preface, “The rules of syntax contained in this work result directly from the analysis of propositions, and of compound sentences; and for this reason the student should make himself perfectly familiar with the sections relating to subject and predicate, and should be able readily to analyze sentences, whether simple or compound, and to explain their structure and connection. * * * This exercise should always precede the more minute and subsidiary labor of parsing.  If the latter be conducted, as it often is, independently of previous analysis, the principal advantage to be derived from the study of language, as an intellectual exercise, will inevitably be lost.”—­Latin Grammar of Andrews and Stoddard, p. vi.  N. Butler, who bestows upon this subject about a dozen duodecimo pages, says in his preface, “The rules for the analysis of sentences, which is a very useful and interesting exercise, have been taken from Andrews’ and Stoddard’s Latin Grammar, some changes and additions being made.”—­Butler’s Practical Gram., p. iv.[332]

OBS. 5.—­Wells, in the early copies of his School Grammar, as has been hinted, adopted a method of analysis similar to the Second one prescribed above; yet referred, even from the first, to “Andrews and Stoddard’s Latin Grammar,” and to “De Sacy’s General Grammar,” as if these were authorities for what he then inculcated.  Subsequently, he changed his scheme, from that of Parts Principal and Adjuncts, to one of Subjects and Predicates, “either grammatical or logical,” also “either simple or compound;”—­to one resembling Andrews and Stoddard’s, yet differing from it, often, as to what constitutes a “grammatical predicate;”—­to one resenbling [sic—­KTH] the Third Method above, yet differing from it, (as does Andrews and Stoddard’s,) in taking the logical subject and predicate before the grammatical.  “The chapter on Analysis,” said he then, “has been Revised and enlarged with great care, and will be found to embody all the most important principles on this subject [.] which are contained in the works of De Sacy, Andrews and Stoddard, Kuehner, Crosby, and Crane.  It is gratifying to observe that the attention of teachers is now so generally directed to this important mode of investigating the structure of our language, in connection with the ordinary exercises of etymological and syntactical parsing.”—­Wells’s School Gram., New Ed., 1850, p. iv.

OBS. 6.—­In view of the fact, that Wells’s chief mode of sentential analysis had just undergone an almost total metamorphosis, a change plausible perhaps, but of doubtful utility,—­that, up to the date of the words just cited, and afterwards, so far and so long as any copies of his early “Thousands” remain in use, the author himself has earnestly directed attention to a method which he now means henceforth