It was recently suggested by Lockwood et al. (2000, 2001) that the cosmic ray intensity in the neutron monitor energy range is linearly related to the coronal source flux, and can be reconstructed for much earlier times using the estimated long term coronal flux. Here we show that a linear regression is oversimplified and leads to unphysical re-sults on long time scales. In particular, the reconstructed cos-mic ray intensity has a steep trend which is four times larger than the allowed upper bound. The reconstructed cosmic ray intensity exceeds the local interstellar cosmic ray flux around 1900. Also, the 11-year cycle minimum of the reconstructed cosmic ray intensity in early 1900s is higher than the highest measured maximum in 1965. We argue that the unphysical results using a linear assumption are due to the oversimplified approach which does not account for complexity and signifi-cant nonlinearity of CR modulation in the heliosphere.