Mixed-make locomotive consists were not anticipated by either the builders or the railroads in the 1940's and early 1950's, so there was no universal m-u system. There were challenges in m-uing some first-generation diesels (F7, RS3, PA1, GP9, etc).
When EMD and Alco-GE units of that era were equipped with dynamic braking, different control systems prevented fully-functional multiple unit operation between the two. EMD used a field loop for DB control, while Alco-GE used potential line control for DB. The field loop required that there be an additional cable, consisting of 3 (or was it 4?) small cables grouped into rectangular plug ends, which were inserted into rectangular receptacles on the ends of the EMD locomotives. This is most easily seen on photos of the exposed receptacles mounted next to the end crossover platforms of GP7/9 and SD7/9 locomotives taken in the 1950's or early 1960's. This connection had to be "looped" by setting a switch on the rear unit to send the control signal back to the controlling unit. Since it was a closed loop, all units in the consist needed to be wired into the loop and all cable connections needed to be tight.
Alco-GE's potential line control simply used wires inside the 27-wire m-u jumper cable.
So, you can see that, when mixed first generation consists were m-ued, the controlling unit could only control dynamic braking back through the consist until it came to the first locomotive of the other manufacturer. The first unit of the other manufacturer - and all units trailing it in the consist - would not have operative dynamic braking.
EMD adopted potential line control beginning with the GP18/20 and SD18/24, and many railroads modified their first-generation EMD's, changing them from field loop to potential line control, and there were no longer any dynamic braking challenges within mixed consists. I remember seeing a switch on a Southern Pacific SD35 which could be positioned either for field loop or potential line control, so the locomotive had evidently been ordered with an option to control dynamic braking on trailing units of either type.
On railroads which did not have dynamic brake-equipped units, m-uing mixed consists was not challenging.
Even among EMD diesels there was some incompatibility. Pre-war diesels such as the E6 and FT used a 21-wire jumper cable and receptacle, while postwar units (E7, F3 and later) used the same 27-wire cable as Alco-GE. There fore, special jumper cables had to be used to m-u E3 and E6 units with E7 and E8 units until the railroad modified the pre-war units.
Another challenge was sanding control. Some Alco-GE units and F-M Erie-Builts had electro-pneumatic sanding control. EMD engines and other Alco-GE's had straight air control. The electro-pneumatic units did not require the additional m-u hose connection on each side of the unit, because they used the control jumper cable to activate sanding on trailing units. This is why you will never see photos of Santa Fe PA's and EMD's m-ued, whereas roads like MP which did not have electro-pneumatic sanding on their PA's freely m-ued them with EMD E-units and passenger GP7's. This did not pose much of a challenge until the transition era between first- and second-generation locomotives. Beginning in the late 1950's (SD24's and RSD15's on my home road), electro-pneumatic sanding became universal. Therefore most railroads converted their older units. Others, like Santa Fe did not, and therefore they did not m-u first- and second-generation units.
I hope this has added enough confusion. 
