In the above paper [1], the author claimed that the similarity solution exists only when the exponent m= [1]. It is shown that this condition insufficient to support his claim. It is found that the similarity solution exists in two cases only : 3/1(i) m= and the slip parameter 3/10=λ(no-slip surface condition); (ii) n=1(Newtonian fluid) and m=1 (linear stretching surface). Also, it is shown that the local skin –friction coefficient defined by the author in ref. [1] is wrong. Consequently, the viscosity parameter α, the radiation parameter R and the Prandtl number have opposite effects on the local skin –friction coefficient than that obtained in ref. [1] .