102 HISTORY OF GREECE. Of tlie late tactic writers, Leo (Tact. vi. 39) and Constantine Por- jiliyroijonitus, repeat the double measurement of the sarissa as given by Polybius. Arrian (Tact. c. 12) and Polyaenus (ii. 29, 2) state its length at sixteen cubits — iElian (Tact. c. 14) gives fourteen cubits. All these authors follow either Polybius, or some other autliority con- current with him. None of them contradict him, though none state the case s>D clearly as he does. ^lessra. Riistow and Kdchly (Gesch. des Griech. Kriegsweseus, (p. 238), authors of the best work that I know respecting ancient military matters, reject the authority of Polybius as it here stands. They maintain that the passage must be corrupt, and that Polybius must have meant to say that the sarissa was sixteen feet in length — not sixteen cubits. I cannot subscribe to their opinion, nor do I think that their criticism on Polybius is a just one. First, they reason as If Polybius had said that the sarissa of actual service was sixteen cubits long. Computing the weight of such a weapon from the thickness required in the shaft, they pronounce that it would be unmanageable. But Polybius gives the actual length as only four- teen cubits: a very material difference. If we accept the hypothesis of these authors — that corruption of the text has made us read cubits where we ought to have read feet, — It will follow that the length of the sarissa, as given by Polybius, would be fourteen feet, not sixteen feet. Now this length Is not sufficient to justify various passages In which Its prodigious length Is set forth. Next, they Impute to Polybius a contradiction in saying that tho Roman soldier occupied a space of three feet. ef|ual to that occupied by a Macedonian soldier — and yet that In the fight, he had two Ma- cedonian soldiers and ten pikes opposed to him (xviil. 13). But there is here no contradiction at all : for Polybius expi-essly says that the Roman, though occupying thi'ee feet when the legion was drawn up in order, required, when fighting, an expansion of the ranks and an in- creased interval to the extent of three feet behind him and on each side of him ('j[(xXa6fia x«t diuaTaaiv aXh'tXcop I'/f/v dti](jti rov^' iiv- dQag hXayiazov TQEig nodag -/.az iniaxariiv y.al nanaoTuzip-) in order to allow full play for his sword and shield. It Is therefore per- fectly true that each Roman soldier, when actually marching up to attack the phalanx, occupied as much ground as two phalangites, and had ten pikes to deal with. Farther, It is impossible to suppose that Polybius, In speaking of cu- bits, really meant feet; because (cap. 12) he speaks of three feet as the interval between each rank in the file, and these three feet are clearly made equal to tico cubits. His computation will not come right, If In place of cubits you substitute feet. We must therefore take the assertion of Polybius as we find it : that the pike of the phalanglte was fourteen cubits or twenty-one feot ill length. Now Polybius had every means of being well Intbrmed on
