[304] This occurrence is mentioned by Seneca, Nat. Quæst. i. 2; he enters into a detailed explanation of the cause; also by V. Paterculus, ii. 59, and by Jul. Obsequens, cap. 128. We can scarcely doubt of the reality of the occurrence, as these authors would not have ventured to relate what, if not true, might have been so easily contradicted. [305] The term here employed is “arcus,” which is a portion only of a circle or “orbis.” But if we suppose that the sun was near the horizon, a portion only of the halo would be visible, or the condition of the atmosphere adapted for forming the halo might exist in one part only, so that a portion of the halo only would be obscured. [306] The dimness or paleness of the sun, which is stated by various writers to have occurred at the time of Cæsar’s death, it is unnecessary to remark, was a phænomenon totally different from an eclipse, and depending on a totally different cause. [307] Aristotle, Meteor. lib. iii. cap. 2. p. 575, cap. 6. p. 582, 583, and Seneca, Quæst. Nat. lib. i. § 11, describe these appearances under the title which has been retained by the moderns of παρήλια. Aristotle remarks on their cause as depending on the refraction (ἀνάκλασις) of the sun’s rays. He extends the remark to the production of halos (ἅλως) and the rainbow, _ubi supra_. [308] This occurrence is referred to by Livy, xli. 21. [309] This meteor has been named παρασελήνη; they are supposed to depend upon the same cause with the Parhelia. A phænomenon of this description is mentioned by Jul. Obsequens, cap. 92, and by Plutarch, in Marcellus, ii. 360. In Shakspeare’s King John the death of Prince Arthur is said to have been followed by the ominous appearance of five moons. [310] This phænomenon must be referred to the aurora borealis. See Livy, xxviii. 11. and xxix. 14. [311] “clypei.” [312] Probably an aërolite. Jul. Obsequens describes a meteor as “orbis clypei similis,” which was seen to pass from west to east, cap. 105. [313] “ceu nubilo die.” [314] It would be difficult to reconcile this phænomenon with any acknowledged atmospherical phenomenon. [315] Perhaps the phænomena here alluded to ought to be referred to some electric action; but they are stated too generally to admit of our forming more than a conjecture on the subject. Virgil refers to the occurrence of storms of wind after the appearance of a falling star; Geor. i. 265-6. [316] These phænomena are admitted to be electrical; they are referred to by Seneca, Nat. Quæst. i. 1. This appearance is noticed as of frequent occurrence in the Mediterranean, where it is named the fire of St. Elmo; see Hardouin in Lemaire, i. 311, and Fouché in Ajasson, ii. 382. [317] Perhaps this opinion may be maintained on the principle, that, when there is a single luminous appearance only, it depends upon the discharge of a quantity of electrical fluid in a condensed state; its effects are, in this case, those that would follow from a stroke of lightning.