We discuss systematics of the M 1 scissors mode within the interacting boson model when the g-boson degree of freedom is included explicitly and microscopically motivated choices of model parameters are adopted. We try to relate the M 1 centroid energy to the energetics of deformation. We conclude that, with the introduction of a hexadecapole-hexadecapole interaction and a g-boson admixture in the ground state of only a few percent, we can obtain reasonable estimates of the M 1 centroid energy, without invoking a Majorana interaction. If one takes seriously variations in microscopic estimates of boson g factors, then the summed M 1 strength near midshell can be interpreted in terms of boson occupation numbers which saturate. ͓S0556-2813͑96͒05206-5͔ PACS number͑s͒: 21.10. Re, 21.30.Fe, 21.60.Fw One of the triumphs of the interacting boson model ͑IBM͒ is its ability to account for the properties of the M 1 scissors mode discovered after its introduction ͓1͔. There are, however, disquieting features: First, there is the fact that to reproduce the excitation energies of 1 ϩ scissors states the somewhat artificial Majorana interaction is apparently required ͓2͔; second, the summed strength does not appear to be consistent with the saturation of the groundstate d-boson occupation number expected near midshell ͓3͔ ͑the ''M 1 saturation'' problem͒. In this paper, we present possible resolutions to these problems. Our primary result is that, if one includes a g-boson degree of freedom ͑in addition to the usual s and d bosons͒, then one can dispense with the use of the Majorana interaction while still adhering to microscopically motivated values of the model parameters. If one takes seriously variations in microscopic estimates of boson g factors near midshell, then the summed strength does admit interpretation in terms of boson occupation numbers which saturate.In earlier work ͓4͔, we identified a deformation contribution E c def to the centroid energy arising from the dependence in the action of the ͑standard͒ quadrupole interaction ϪQ p •Q n on the neutron-proton ͑or F-spin͒ symmetry of states. The magnitude of this deformation contribution would have been inadvertantly underestimated in ͓2͔ because of the nonstandard ͑and microscopically implausible͒ F-spin scalar quadrupole interaction Ϫ(Q p ϩQ n )•(Q p ϩQ n ) adopted. For the deformed Sm isotopes, we found that E c def almost certainly could account for a substantial fraction of the centroid energy ͑80% or so for the choice of model parameters made in ͓4͔͒. In this paper, we attempt to understand the energetics of the M 1 scissors state solely in terms of the deformation contribution E c def . We believe this approach to be natural: Within the sdIBM-2, the M 1 scissors mode may be viewed in the intrinsic frame as an F-spin isovector quad-rupole excitation ͓5͔; within the sdgIBM-2 ͑which we consider below͒, the scissors mode is a superposition of F-spin isovector quadrupole and hexadecapole excitations.In most IBM studies of the M 1 scissors mode, it is customary t...