DISCUSSION

DR. MANDEL: Prof. O'Konski, does the average relaxation time, which you measured for the superhelical structure, really correspond to the rotational relaxation time? DR.O'KoNSKI: In terms of a flexible model, yes. That is, the times are within the range that I would expect. If you would like a segmental model, one way of looking at it is to consider that there are segments of various lengths and inertial effects are negligible, so you can calculate the frictional coefficient of rodlike segments and from that you can get a distribution of relaxation time. If you look at the shortest relaxation time, about ten microseconds, that would correspond to a length of about 2,000 Afor the shortest persistence length. Now that corresponds quite nicely with the expectation of a persistence length on the order of a couple of thousand angstrom for DNA. That also raises a point with respect to Prof. Houssier's calculation using your model where the discrepancy between the persistence lengths from this measurement and from others and B values he calculates is quite appreciable. DR. MANDEL: One has to be very careful if one uses persistence length, because I think the wormlike chain is only well defined at the dilution. DR.O'KoNSKI: We are at a very low concentration of the DNA, down to the order of micrograms per mil, and in that region even with the longer calf thymus DNA, the relaxation times go asymptotically to a constant value. DR. MANDEL: If the molecules are very large, one has to be very careful to define what infinite dilution is. One parameter we can use is the overlap volume, supposing every molecule must turn more or less freely without touching each other at infinite dilution. That should define the region of infinite dilution, and I have the feeling that with very large molecules, this low concentration where no overlap occurs must be really low. DR. O'KONSKI: I have done these calculations years ago, and I am confident that we are at the essentially infinite dilution value for the PM-2 DNA. DR. HOUSSIER: We should not compare the values that were presented by Professor Mandel from the dielectric dispersion and the one I presented from electric birefringence, for two reasons. First of all, the ionic strength was not the same in the two measurements. Second we are discussing average values and the type of averages that we are obtaining from relaxation processes, and from electric polarizability measure-