Confusions on Pair_Coeff in Gayberne Potential

Sorry for sending out the incomplete one by mis-pressed the send button.

Dear Lammps users,

I am writing to make sure that my understanding in pair_coeff is correct.
My confusion lies in here:

For better understanding,I listed these values in a table form. (The information below is taken from Lammps default input file of ellipse)

type i type j epsilon sigma epsilon_i_a epsilon_i_b epsilon__c epsilon_J_a epsilon_J_b epsilon_J_c

pair_coeff 1 1 3.0 1.0 1 1 1 1 1 1
pair_coeff 1 2 3.0 1.0 1 1 1 0 0 0
pair_coeff 2 2 1.0 1.0 1 1 0. 2 0 0 0

According to the Lammps description of values of these epsilons, where it mentioned ’ the typical way to define the epsilon_i and epsilon_j coefficients is to list their values in “pair_coeff I J” commands when I = J, but set them to 0.0 when I != J.', and then here comes my question.

If say pair_coeff for type 1 to 1 is supposed to be listed (b/c list values when I=J),especially epsilon_J_(a,b,c), then why for type 2 to 2 (I=J) the pair_coeffs are not listed for epsilon_J(a,b,c)? According to my understanding after reading the manual, the epsilon J_a, epsilon J_b, epsilon J_c should be assigned values as well.

Can someone kindly explain to me how are these values supposed to be assigned?
Thanks in advance for any ideas!