With the more extensive data from new x-ray films of the diffraction pattern of stretched rubber made in a cylindrical camera, an attempt has been made to find the true unit cell. No really satisfactory solution has been reached. The leading possibilities are described. They are characterized by a change of the angle between the two independent sets of planes which reflect the two strong equatorial spots from 90°, the value which first suggests itself, to 109.5°. The new choice is shown to be consistent with all the evidence. It has the great advantage of resolving fairly well the familiar discrepancy between measured and calculated values of the density of the crystals in stretched rubber. The new value from the unit cell is 1.00. A necessary consequence of the x-ray data is that the carbon atoms of a molecular chain are not co-planar in stretched rubber. The reason appears to be the large repulsive forces which would have to exist in that configuration. In rotations on the single bonds during stretching and retraction the interatomic forces are probably repulsions in the dominating pairs. Repulsions give a plausible explanation of the form of the outer part of the stress-strain curve.