Bordman felt even deeper depression as he worked out the facts. His proposal was essentially futile. It would be encouraging, and to a very slight degree and for a certain short time it would palliate the situation on the inner planet. But in the long run its effect would be zero. He was embarrassed, too, that Herndon was so admiring. Herndon would tell Riki that he was marvelous. She might--though cagily--be inclined to agree. But he wasn't marvelous. This trick of a flier-supported grid was not new. It had been used on Saril to supply power for giant peristaltic pumps emptying a polder that had been formed inside a ring of indifferently upraised islands. _All I know_, thought Bordman bitterly, _is what somebody's showed me or I've read in books. And nobody's showed or written how to handle a thing like this!_ He went to Herndon's desk. Herndon had made a new graph of the solar-constant observations forwarded from home. It was a strictly typical curve of the results of coinciding cyclic change. It was the curve of a series of frequencies at the moment when they were all precisely in phase. From this much one could extrapolate and compute. Bordman took a pencil, frowning. His fingers clumsily formed equations and solved them. The result was just about as bad as it could be. The change in brightness of the sun Lani would not be enough to be observed on Kent IV, the nearest other inhabited world, when the light reached there four years from now. Lani would never be classed as a variable star, because the total change in light and heat would be relatively minute. The formula for computing planetary temperatures is not simple. Among its factors are squares and cubes of the variables. Worse, the heat radiated from a sun's photosphere varies not as the square or cube, but as the fourth power of its absolute temperature. Bordman's computations were not pure theory. The data came from Sol itself, where alone in the galaxy there had been daily solar-constant measurements for three hundred years. The rest of his deductions were based ultimately on Earth observations, too. Most scientific data had to refer back to Earth to get an adequate continuity. And there could be no possible doubt about the sunspot data, because Sol and Lani were of the same type and nearly equal size. Using the figures on the present situation, Bordman reluctantly arrived at the fact that here, on this already-frozen world, the temperature would drop gradually until CO_{2} froze out of the atmosphere. When that happened, the temperature would plummet until there was no really significant difference between it and that of empty space. It is carbon dioxide which is responsible for the greenhouse effect, by which a planet is in thermal equilibrium only at a temperature above its surroundings, as a greenhouse in sunlight is warmer than the outside air. The greenhouse effect would vanish soon on the colony-world. When it vanished on the mother planet....