I will discuss the processes through which three dimensional turbulence develops in stratified free shear layers. In particular, I will focus on the impact of various secondary instabilities on the mechanism of transition to turbulence. Recent observations and numerical simulations have demonstrated the possibility of existence of new secondary shear instabilities developing on the primary Kelvin Helmholtz billows. In this study, I investigate the competition between the newly-found and well-known secondary instabilities for control of the transition process and provide probability maps for their regions of dominance in parameter space. I will try to address three important issues: First, I will argue that there are critical differences between laboratory experiments on shear layers and their geophysical counterparts, second, I will show that an upscale cascade (through vortex pairing), which is typically observed in laboratory setups, gets suppressed once one moves to regions in the flow parameter space more like geophysical flows, and third I will show that differences between mechanisms responsible for turbulence collapse of a stratified layer have significant implications for the 'Mixing efficiency', a parameter which once defined properly, can be used for parametrization of sub-grid shear mixing in Ocean General Circulation Models. The last two points have implications for the general topic of turbulence in stratified flows.