Return of the habitats constructed from the 104 additional asteroids will now be discussed briefly, for the case of habitat construction without solar power satellites and construction in low earth orbit. For some of these asteroids, the travel time from the first asteroid will be about the same as in the case of habitat construction with solar power satellites, or 8 months, and the return time the same 51 months. There are 15 generations constructed on the time the same 51 months, first asteroid before the 104 space manufacturing facilities and mass drivers are constructed. Since the construction takes 3 months, the arrival of the first of the additional habitats in their final high earth orbit is (15 • 3) + 3 + 8 = 56 months after the habitats from the first asteroid, or at month 161 - 13 + 56 = 204 = 17 yr. Discussion The cost estimates are considered uncertain by 50%, with the exception of the cost to develop the self-reproducing factory, which is less certain. It is possible, though unknown, that S.S.I. assumed a separately funded program to develop a self-reproducing factory, and that this would be followed by the development of the particular manufacturing facilities used for this plan. The cost of $10$ /t for the most intensively researched component of these facilities would then be only that for the adaptation required for the development and construction of these particular facilities. S.S.I. did not give the derivation of the mass which requires research at this cost, so to adapt this to the other plans in this paper, the mass which requires this research in the plans in the present paper which use self-reproduction was determined by estimating that the same proportion of the total mass as in S.S.I. requires this research. This assumption was made because no other information was available, and is too crude to yield an acceptably reliable estimate. No matter how accurately this proportion is determined, however, the assumption that costs can be estimated by assigning a certain cost per unit mass to research, and construction, of equipment is unacceptably unreliable. This is especially true for the facilities in S.S.I., since this equipment would be more advanced than any in existence; only a detailed design is reliable enough for the estimation of costs. Also, manufacturing facility mass estimates are not reliable for the same reason, that detailed designs would be required to determine them. S.S.I. was presented three months after the speech by G. K. O'Neill at Stanford in January 1980, in
RkJQdWJsaXNoZXIy MTU5NjU0Mg==