30 TRANSACTIONS OP THE 



prevails in any of the languages under consideration, with the 

 exception of Italian. 



Reference was incidentally made to deviations by classic writers 

 and reputable modern Latinists from the rule that " dependent 

 clauses, containing an indirect question, take the subjunctive." It 

 was shown that, while such deviations were frequent in Plautus and 

 Terence, they were of by no means rare occurrence in the writings 

 of many Latin poets, of Seneca, and even of Cicero, although, in 

 the instances in which the last-named writer deviates from this rule, 

 modern editors have made him conform to it. 



Attention was called, in conclusion, to certain points connected 

 with the use of the tenses of the Portuguese subjunctive, in refer- 

 ence to which modern grammarians continue to advocate a usage 

 which is really, at the present day, practically obsolete. 



The paper was discussed by Prof. Fay, Dr. Antisell, Dr. Welling, 

 Mr. Ward, and Prof. Mason. 



Prof. Fay expressed his gratification at listening to so learned a 

 production, and regarded it as confirmatory of the views he had 

 previously presented to the Society. 1 



He regretted that it was not possible to systematize the data in 

 more exact chronological order. 



Dr. Antisell remarked upon the tendency of languages to dis- 

 pense more and more with the subjunctive mood as to a great 

 extent a useless appendage. 



Dr. Welling also noted the progress made in sloughing off 

 redundant forms, and thought it quite possible that this process 

 might go too far. He said that there was a very perceptible differ- 

 ence in the meaning conveyed by the English subjunctive and 

 indicative, which it would be a pity to lose the power of expressing. 



Mr. Ward was interested in the cases adduced in Latin of the 

 dependent interrogative with the indicative, which appears to be 

 chiefly used in colloquialisms, showing that the common people 

 were disposed to eliminate unnecessary grammatical formalities. 



1 See p. 26, supra. 



