17.  We cannot strictly speak at all of a fixed number of senators.  Though the censors before Sulla prepared on each occasion a list of 300 persons, there always fell to be added to this list those non-senators who filled a curule office between the time when the list was drawn up and the preparation of the next one; and after Sulla there were as many senators as there were surviving quaestorians But it may be probably assumed that Sulla meant to bring the senate up to 500 or 600 members; and this number results, if we assume that 20 new members, at an average age of 30, were admitted annually, and we estimate the average duration of the senatorial dignity at from 25 to 30 years.  At a numerously attended sitting of the senate in Cicero’s time 417 members were present.

18.  II.  III.  The Senate.  Its Composition

19.  IV.  VI.  Political Projects of Marius

20.  III.  XI.  Interference of the Community in War and Administration

21.  IV.  VII.  Legislation of Sulla

22.  II.  III.  Restrictions As to the Accumulation and the Reoccupation of Offices

23.  IV.  II.  Attempts at Reform

24.  To this the words of Lepidus in Sallust (Hist. i. 41, 11 Dietsch) refer:  -populus Romanus excitus... iure agitandi-, to which Tacitus (Ann. iii. 27) alludes:  -statim turbidis Lepidi rogationibus neque multo post tribunis reddita licentia quoquo vellent populum agitandi-.  That the tribunes did not altogether lose the right of discussing matters with the people is shown by Cic.  De Leg. iii. 4, 10 and more clearly by the -plebiscitum de Thermensibus-, which however in the opening formula also designates itself as issued -de senatus sententia-.  That the consuls on the other hand could under the Sullan arrangements submit proposals to the people without a previous resolution of the senate, is shown not only by the silence of the authorities, but also by the course of the revolutions of 667 and 676, whose leaders for this very reason were not tribunes but consuls.  Accordingly we find at this period consular laws upon secondary questions of administration, such as the corn law of 681, for which at other times we should have certainly found -plebiscita-.

25.  II.  III.  Influence of the Elections

26.  IV.  II.  Vote by Ballot

27.  For this hypothesis there is no other proof, except that the Italian Celt-land was as decidedly not a province—­in the sense in which the word signifies a definite district administered by a governor annually changed—­in the earlier times, as it certainly was one in the time of Caesar (comp.  Licin. p. 39; -data erat et Sullae provincia Gallia Cisalpina-).