From (1) and (2_a_) he concluded that the system prevalent in his time of appointing one of the strategi as superior to his colleagues in the command of some special expedition or department prevailed at the time of Marathon; that is to say, that Miltiades was in a position analogous to that which Perikles had held on several occasions in Herodotus’ own time. Still, he is evidently uncertain on this point; and he accounts for Miltiades’ position by assuming a system of diurnal rotation in the command, ◆H. vi. 110.◆ and by representing that some of Miltiades’ colleagues surrendered their days of command to him. From (2_b_) he draws the very important conclusion that the polemarch was no longer commander-in-chief. He has ante-dated this particular application of the system of lot by three years. It is obvious that the Athenian citizen would not entrust his life when serving in the field to a commander appointed by such a method. His belief in Providence, as represented by the lot, had its limitations. Herodotus was probably well aware that, from the moment of the introduction of the lot into the election of the archons, the polemarch had ceased to command the army. Perhaps, after all, the main difficulty which would be experienced by an inquirer in the latter half of the fifth century into the system of command existent in 490, would arise from the fact that that year fell within what was a comparatively brief period of transition in the Athenian army system, intermediate between the old organization existent before the reforms of Kleisthenes, and the completely new organization which was introduced between 490 and 480, probably in 487, and prevailed with hardly any modification during the rest of the century. The reforms of Kleisthenes, by substituting ten tribes for four, had rendered a certain amount of army reorganization necessary. Thereafter the army was constituted in ten tribal regiments, each commanded by one of the strategi:[75] “they chose the generals by tribes, one from each tribe.” What other _novelties_ there were in the system we do not know; but of the system itself, as existent at Marathon, we derive a certain amount of knowledge from the Aristotelian treatise on the Constitution of Athens, and with its aid we may now proceed to set in order the confused and mistaken account of Herodotus. That treatise distinctly states (Ch. 22) that the polemarch was chief-in-command at Marathon. This is, perhaps, a conclusion at which a close student of Herodotus’ uncertain account would arrive. ◆H. vi. 109.◆ The attitude of Miltiades towards him at the celebrated Council of War is that of an official inferior. His command of the right wing would argue that he was, at least _de jure_, the chief of the army. Of the position of the strategi the treatise says but little. It seems certain, however, that, in addition to commanding the tribal contingents, they formed, as Herodotus implies, a council of war presided over by the polemarch.