An angle of two seconds is that in which a circle 610 of an inch in diameter would be seen at the distance of a mile. This is the star which shows the greatest parallax of all. The parallax of the bright star of Lyra is not more than a quarter of a second. Struve, at the Observatory of St. Petersburgh, has deduced, as he thinks, from observations, that for stars of the second magnitude the general average of parallax is 110 of a second. This is so small an angle that it is almost impossible to answer for it. Supposing, however that it is 110 of a second, then the distance of the star from the sun is two million times as great as the distance of the earth from the sun. It seems almost inconceivable that we should be able to measure, even in a rough way, a distance so great.
I will only mention one more thing. There is one correction upon which I said there was a little doubt, and that is that troublesome thing, refraction. It is one of those things which throw a doubt upon every observation of a delicate kind. Refraction enters here, because we must necessarily observe the zenith-distance of the star; and in comparing observations of zenith-distance at opposite times of the year, there is this unfortunate circumstance: the same star which is observed on the meridian in the day-time at winter, will be seen on the meridian at night in the summer; or the star which is observed in the night in the winter, will be observed in the day-time in the summer, when the state of the air is very different; so that the amount of refraction at the two observations will be very different, and we cannot determine the correction to the zenith-distance accurately, so as to answer for 110 of a second between the observations. Under these circumstances, this determination of a difference between the observations at E and E'",
