Phase portraits of two classes of quadratic differential systems exhibiting as solutions two cubic algebraic curves (Q6045558)

The authors want to study the phase portraits of two families of quadratic systems which have a concrete invariant conic. There is not much problem in their first system (2) since it is just a one parameter family with an invariant stright line which produces only two different phase portraits. But the second family (3) has 4 parameters, and even if it is true that it always has the desired invariant curve, the results given are partially wrong and clearly incomplete. I fear that the authors are not aware of the difficulties of studying such a large family. Even though this is not a referee report, I have found several mistakes in phase portraits from 3 to 7, and two missing phase portraits among the conditions which should produce those first pictures. A family with so many parameters cannot be easily studied with the classical tools of locating singular points and studying their Jacobians. In these cases, the use of invariants has proved to be much more powerful. And I also doubt that this study can be done in a completely algebraic way. From some inconsistencies I detect between phase portraits 6 and 7, I fear that there may appear also nonalgebraic bifurcations. My recommendation to the authors is that they should start to use these much more powerful tools and make a new work on this family which promises to be an interesting one.