His discussion of the rainbow (16–33) is one of the most detailed and vivacious in the whole volume. It takes the form of a sustained argument, in which the author cites various authorities, and replies to objections brought by a supposed opponent to his thesis, which is that the rainbow is unquestionably an image of the sun received in a very moist cloud which has the shape of a round concave mirror (20, 27). He quotes with apparent approbation the opinion that in a shower of rain each falling drop is a mirror reflecting an image of the sun, and that when an observer stands directly between the sun and the shower he sees the reflections of the countless drops blended into one continuous semicircle. But as the discussion proceeds the writer denies that the cloud consists of separate rain-drops, and he maintains that even if it did they would not unite to give one unbroken image. In proof of his contention he urges the fallacious assertion that if a number of mirrors are joined together and a man is placed before them, each gives its own reflection, and thus a single man becomes multiplied into a crowd. If he had ever tried the experiment or had visited the shop of a mender of mirrors, he would have seen that the separate pieces, if strictly arranged on the same plane, reflect a single image. His imaginary antagonist asks for an explanation of the rainbow-like colours displayed by the spray from a burst water-pipe, or the splash from an oar, which are, of course, cases strictly parallel to the falling shower of rain (24). The resemblance is at once granted, but is explained away on the ground that the drops fall so quickly that they cannot form reflections of the sun, and that to produce such reflections the medium must be at rest. The objector once more strikes in with a reference to the rainbow colours to be seen in a glass rod which is placed obliquely in the path of the sun’s rays (30). These prismatic tints, as has long been known, are due to the same decomposition of white light, as in the rainbow. But Seneca claims the illustration as furnishing additional arguments in his favour. He maintains that no colour is really produced in the rod, but only a false appearance of colour, his idea being apparently that unless the colour is inherent in an object apart from direct sunlight, it is only apparent and not real. The glass, he says, tries to reproduce the sun’s image, but fails because of its unsymmetrical form, the reflections being crowded together and confused into the appearance of a single band of colour. In regard to the falling drops of rain in a shower he contends that they receive the colour but not the image of the sun, and he is led away by the false analogy of the varying tints of a peacock’s neck as the bird tosses its head (25). At one part of the discussion he affirms that the colours of the rainbow come partly from the sun and partly from the moist cloud (21). Further on, however, he agrees