must mean that, at any given moment, half of the sea is taking the downward path, and has just been fiery storm-cloud, while half of it is going up, and has just been earth. In proportion as the sea is increased by rain, water passes into earth; in proportion as the sea is diminished by evaporation, it is fed by the earth. Lastly, the ignition of the bright vapour from the sea in the bowl of the sun completes the circle of the “upward and downward path.” Footnote 379: This was written in 1890. In his _Herakleitos von Ephesos_ (1901) Diels takes it as I did, rendering _Glutwind_. Footnote 380: Cf. Herod. vii. 42, and Lucretius, vi. 424. Seneca (_Quaest. Nat._ ii. 56) calls it _igneus turbo_. The opinions of early philosophers on these phenomena are collected in Aetios, iii. 3. The πρηστήρ of Anaximander (Chap. I. p. 69, _n._ 133) is a different thing altogether, but it is quite likely that Greek sailors named the meteorological phenomenon after the familiar bellows of the smith. Footnote 381: Aet. iii. 3, 9, πρηστῆρας δὲ κατὰ νεφῶν ἐμπρήσεις καὶ σβέσεις (sc. Ἡράκλειτος ἀποφαίνεται γίγνεσθαι). Diels (_Herakleitos_, p. v.) seems to regard the πρηστήρ as the form in which water ascends to heaven. But the Greeks were well aware that waterspouts burst and come down. [Sidenote: Measure for measure.] 72. The question now arises, How is it that, in spite of this constant flux, things appear relatively stable? The answer of Herakleitos was that it is owing to the observance of the “measures,” in virtue of which the aggregate bulk of each form of matter in the long run remains the same, though its substance is constantly changing. Certain “measures” of the “ever-living fire” are always being kindled, while like “measures” are always going out (fr. 20); and these measures the sun will not exceed. All things are “exchanged” for fire and fire for all things (fr. 22), and this implies that for everything it takes, fire will give as much. “The sun will not exceed his measures” (fr. 29). And yet the “measures” are not to be regarded as absolutely fixed. We gather from the passage of Diogenes quoted above that Theophrastos spoke of an alternate preponderance of the bright and dark exhalations, and Aristotle speaks of Herakleitos as explaining all things by evaporation.[382] In particular, the alternation of day and night, summer and winter, were accounted for in this way. Now, in a passage of the pseudo-Hippokratean treatise Περὶ διαίτης which is almost certainly of Herakleitean origin,[383] we read of an “advance of fire and water” in connexion with day and night and the courses of the sun and moon.[384] In fr. 26, again, we read of fire “advancing,” and all these things seem to be intimately connected. We must therefore try to see whether there is anything in the remaining fragments that bears upon the subject. Footnote 382: Arist. _de An._ Β, 2. 405 a 26, τὴν ἀναθυμίασιν ἐξ ἧς τἆλλα συνίστησιν. Footnote 383: