George F. Mull, Professor of Latin, Franklin and Marshall College. Oscar D. Robinson, Principal of the High School, Albany, N. Y. Joseph C. Rockwell, Assistant Professor of Archaeology, University of California. Leigh Richmond Smith, Teacher of Latin and Greek, High School, San José. Edward M. Traber, Instructor in Latin and Greek, State Agricultural College, Fort Collins, Colo. Milton H. Turk, Professor of English, Hobart College. Elmer E. Wentworth, Professor of English, Vassar College. 2. The TRANSACTIONS and PROCEEDINGS for 1895 (Vol. XXVI) were issued in April. Separate copies of the PROCEEDINGS may be obtained of the Secretary or of the Publishers. 3. The Report of Publications by members of the Association since July 1, 1895, showed a record of books and pamphlets by over fifty-five members. To ensure the completeness of the report, it is earnestly requested that every member enter his publications upon the blanks to be sent out in June of each year. It is desirable that only those publications be entered on the list which have a distinctly philological character. 4. The contract with the publishers, Messrs. Ginn & Co., has been renewed for a term of five years. Professor Smyth then made his report as Treasurer for the year 1895-96 : The reading of papers was then begun. At this time there were present about fifty members. At subsequent meetings over seventy members were in attendance. 1. Children on the Stage in the Sanskrit Drama, by Professor A. V. Williams Jackson, of Columbia University. A motto for the paper was found in Hamlet's allusion to the children players in London in the days of Queen Elizabeth (Ham. ii 2. 330). Other instances of children on the stage in the English drama from its beginning to the time of Shakspere were brought out, and attention was called to the presence of children, as a recognized element in histrionic productions on the Greek stage, and sporadically also in the Latin theatre. The investigation then turned to India. The romantic character of the ancient Hindu plays was first treated of with reference to the free non-observance of the unity of time. The plot of Kālidāsa's Çakuntala was chosen as an illustration of the lapse of time during the progress of a play. The dramatic part which Sarvadamana, the little son of the hero and heroine, plays in the dénouement of this romantic piece (act vii) was emphasized. Also in Kālidāsa's Vikramōrvaçī, the character of Ayus, scion of the king, served as a good example of a youth's bringing about the happy solution of an involved play. Bhavabhūti's Uttara-Rāma-Carita, a sort of Sanskrit Winter's Tale, offered parallels to Shakspere. Kuça and Lava, placed under the guardianship of the sage Valmiki, become striplings of heroic mould like Guiderius and Aviragus reared by old Belarius in Shakspere's Cymbeline; and in the sixth act these manly youths are restored to their father, Rāma. In the interlude, or masque production, which is presented in the last act of the drama (act vii), the circumstances of the birth of the two heroic princes are enacted in mimic reality before the king. In this scene the banished queen appears before the audience, supported on either side by Earth and Ganges. These latter impersonations, as the paper showed from the stage-direction, were intended to be represented as holding each an infant boy in the arms (tataḥ praviçaty utsañgitāi 'kāikadārakhābhyaṁ pṛthvigangābhyām avalambitā sītā). In whatever manner the scene was presented, whether merely by pantomimic gesture or by some more realistic device, none the less, the notion of a child in swaddling clothes is portrayed dramatically, just as in the Winter's Tale (ii. 3) or in Terence's Andria (ii. 6-7). The royal boys whose birth the mimic play enacts, are now grown to be twelve years old, as the play tells us (act iii et al.). Like Ayus of the Vikramõrvaçi, they speak Sanskrit, not Prakrit. As already noted of the Çakuntala, the paper observed that also in the Mṛcchakatika the little Rohasena, son of the hero whose fortune has been ruined, is very young and speaks Prakrit in the climax scene where the lad is introduced. This is the scene, so full of tenderness, that gives the name Toy Cart' to the play. In the last act of the same drama (act x), the little fellow is again brought in to add to the pathetic situation of the last hours of a father unjustly condemned to die. The dramatic character of this scene was criticised with some detail. A parallel situation was cited from Viçākhadatta's Mudrā Rākshasa (act vii), where a child is similarly brought on the stage in the scene of the impending execution of a guiltless father sentenced to death. In this play also the young child speaks the Prakrit dialect. One other instance of a touching rôle played by a child was adduced from Kshemiçvara's Canḍakāuçika. The tiny boy's thoughtless and childish Prakrit prattle me too' (mam pi) adds depth to the heart-rending nature of the scene in which the unfortunate parents are sold into servitude (act iii), and his seeming death and miraculous restoration to life, in the last act (act v), complete the mingled woof and weft of joy and sorrow that make up the material of this noble tragi-comedy. The paper closed by estimating the importance of the rôle played by children in the Sanskrit dramas, as compared with the histrionic productions of other nations; and it favorably criticised the faithfulness of touch and the power of expression in portraying the natural love of children which the Hindu plays showed. The concluding paragraph emphasized several points of interest which the early dramas of India possess in the light of parallels that they offer to the plays of Shakspere. 2. Age at Marriage in the Roman Empire, by Professor Albert Granger Harkness, of Brown University. This paper is printed in full in the Transactions. 3. Notes on the Etymology of Atrium, by Dr. H. W. Magoun, of Oberlin, O. Probably no other word in Latin, or indeed in the classical languages, has had more etymologies proposed for it than the word atrium. No less than seven have been seriously put forward. Two of them are Greek, dopóov, suggested by Becker, and atopiov, once largely accepted, and by general consent attributed to Scaliger, who seems to have been the first to propose it. His contemporary, Casaubon, advocates this same view, and in his notes, animadversiones, p. 99, on Suetonius, Augustus, 29 (p. 37, l. 25, atrium Libertatis), he says: "in medio erat area sub dio, columnis cincta: ideo is locus etiam peristylium appellatur. Idem quoque impluuium dicebatur. Atque hoc proprie atrium est, non atrii pars: nam atrium ab al@piov, significat locum in aedibus sub dio. Graeci napov frequentius vocant: sed atopiov pro atrio notauimus saepe apud LXX. & Iosephum. . . . Extat & apud Lucianum," etc. If these two etymologies presented no phonetic difficulties, there would still remain the question of the historical connection to be solved. It can hardly be supposed that the word came into Latin from the Greek through Tuscan and, if it did not, no other bridge appears until the fourth century B.C., which hardly gives time enough for the Romans to completely forget such an origin, as they inust have done to accord with the facts. In the case of αίθριον, it is clear that its adaptation to the meaning atrium, which from the late date of its appearance in Greek in this sense is plainly only a case of popular etymology from the Greek side, has led to the error of reversing the truth in a sense and supposing that atrium came from αἴθριον. Ottfr. Mueller, Etrus. I. 256, draws a comparison, which might possibly be expressed in the form of an equation, between the Atrias on the Adriatic Sea and the atrium. He says: "Wie der Atrias am adriatischen Meer ursprünglich das Land der zusammenfliessenden Ströme (Athesis, Tartarus, Padus u. s. w.) und der Sammelplatz aller Gewässer Ober-Italiens ist: so ist das Atrium der Theil des Hauses, wo das Wasser, welches auf das Dach herabregnet, im compluvium und impluvium zusammenfliesst." See Beck. Gal.2 II. p. 251. This needs no comment; it is the conception of a poet or a Donnelly. Festus, quoted by Paulus I. 12, gives two alternatives. The first agrees with Varro, cited below, to which he adds: vel quod a terra oriatur, quasi aterrium. This also may be passed over. Isidor. Or. XV. 3, 4, says of it: dictum est atrium, quod addantur ei tres porticus extrinsecus. Aliis atrium quasi, etc., which may well be classed with the etymology proposed by Mueller. The other view which he proceeds to give is the same as that of Servius cited below. Varro, L. L. V. 161, says: Atrium appellatum ab Atriatibus Tuscis; i.e. from the Tuscan town of Atria, a suggestion which is plausible, although Casaubon ridicules it with others, loc. cit.: “Varronis aliorumque veterum notationes quis non rideat?" Varro's view, however, carries with it more than seems probable. If the etymology is correct, the Romans either had no atrium at all or none properly speaking until the Tuscan form of building was adopted. The first supposition is contrary to the natural development of the domus from the casa: in fact all building everywhere seems to have begun with the tent or hut or cave having a single common room to which others were added in the course of time. The second supposition restricts the application of the word originally to the Tuscanicum, which Mau (Marq. Privatl. d. Röm.2 I. p. 223, n. 4) believes to be the meaning of Varro. But this involves both the question of the date of the adop tion of the Tuscan form of building, which Göll (Beck. Gal.2 II. p. 253) thinks may have become general after the burning of Rome by the Gauls (390 B.C.), and the name applied to the original living room of the early Romans before the Tuscanicum became common. It also leaves such expressions as atrium Vestae to be accounted for and presents other minor difficulties. Such questions as these, which may never be finally settled, manifestly cannot be included within the scope of the present paper. Servius, Ad Aen. I. 726, in speaking of the atrium, says: “Ibi et culina erat: unde et atrium dictum est; atrum enim erat ex fumo." Strangely enough, Becker (Gal.2 II. p. 251) says of this etymology: "Servius zu Aen. I, 730 leitet es gar vom Rauche ab: " but Servius plainly gives the word a history similar to that of the Greek μéλa@pov, and by so doing allows a very early origin for it. The extreme probability of the correctness of this view has now led to its general acceptance; see Marq. Privatl.2 I. p. 218. It is, moreover, a curious fact that Varro himself indirectly supports this etymology; for he derives the masculine form from the same stem. He says, L. L. VIII. 451: alia [nomina] a vocabulo ut ab albo Albius, ab atro Atrius. 4. The Problem of the Atriolum or the meaning of the word in Classical Latin, by Dr. H. W. Magoun, of Oberlin, O. The atriolum, so far as has been noted, is mentioned but twice in Classical Latin, and both passages are in Cicero: ad Att. I. 10, 3: praeterea typos tibi mando, quos in tectorio atrioli possim includere, et putealia sigillata duo; and ad Qu. fr. III. 1, 1, 2: quo loco in porticu te scribere aiunt ut atriolum fiat, mihi, ut est, magis placebat. Neque enim satis loci videbatur esse atriolo, neque fere solet nisi in his aedificiis fieri, in quibus est atrium maius, nec habere poterat adiuncta cubicula et eiusmodi membra. The word also occurs in an inscription, a few times in the LXX., and occasionally in the writers of mediaeval times; but these passages do not materially affect the question. Both of the citations occur in the Letters, and it may perhaps be inferred that the usage was colloquial. It appears from Cicero's words that the room was adorned with figures on the walls; had a puteal, cubicula, etc.; and was in short very similar to an ordinary atrium. The natural inference is that it was merely a second atrium of smaller size than the first. Metcalf concludes (Beck. Gal. p. 253), that it served as "an antechamber to a greater hall, peristylium with a porticus," and that the atriola "were only to be found in large mansions." Marquardt finds in Pliny, Ep. II. 17, an explanation of their character, and his position is accepted by Göll (Beck. Gal.2 II. p. 246) as sound. He assumes (Privatleben d. Röm.2 I. p. 223, n. 4) that the D-shaped porticus with their included area formed the peristylium of Pliny's villa, and that the cavaedium hilare must have been identical with the atriolum which Q. Cicero wished to join (anlegen) to the porticus. (This notion that the D-shaped porticoes formed the peristylium was suggested as early as 1832, Lib. of Entertaining Knowl., Pompeii, II. p. 8, footnote, issued by Soc. for Diffus. of Use. Knowl., and the suggestion seems plausible except that peristylia were regularly rectangular in shape, or at least their sides were straight. According to Vitruvius, VI. 4, they should be a third part longer than wide, and be placed transversely.) The resulting villa, if Marquardt be followed, would be of a very unusual construction first, an atrium with its vestibulum; then, a peristylium whose area was very small, parvula, a little bit of a one'; next, a small atrium; and behind this, a triclinium or oecus. Nothing of the sort appears to have been found, and it is doubtful whether it ever will be; for an interior atrium, where there is an exterior one, seems to be an anomaly, and an atriolum is merely a small atrium. Of the twelve conjectural plans of the villa, which I have been fortunate enough to collect (Scamozzi, 1615; Felibien des Avaux, 1699; Castell, 1728; Marquez, 1796; Hirt, 1827; Haudebourt, 1838; Schinkel, 1841; Bouchet, 1852; Burn [after Hirt], 1871; Cowan, 1889; Winnefeld, 1891; and Magoun, 1894) not a single one can be regarded as favoring the view of Marquardt; for in no case is the cavacdium represented as smaller than the atrium, which it must be, to be the atriolum (cf. Cicero's statement), and four, including the two latest, regard the cavaedium hilare as a peristylium. Now it appears that there were porticoes in villas besides those in the peristylium; for Vitruvius, in giving the arrangement of a country villa, says, VI. 8: ruri autem pseudourbanis statim peristylia, deinde tunc atria habentia circum porticus pavimentatas spectantes ad palaestras et ambulationes. Moreover, in the passage from Cicero upon which Marquardt bases his theory, no peristylium is mentioned. In fact the word does not occur in either epistle; but in the preceding section he says: villa mihi valde placuit, propterea quod summam dignitatem pavimentata porticus habebat: and again, in the other passage just preceding the citation above, he says: Signa nostra. . . velim imponas, et si quod aliud olkeîov . . . reperies, et maxime, quae tibi palaestrae gymnasiique videbuntur esse. Etenim ibi sedans haec ad te scribe |